JP3507436B2 - Wireless internet access system - Google Patents

Description

DETAILED DESCRIPTION OF THE INVENTION [0001]Cross-reference with related applications This application is based on U.S. Provisional Patent Application No. 60/08 filed April 17, 1998.
Claims the benefit of priority of 2073
The content of the request is incorporated herein by reference. [0002]Background of the Invention Field of the invention The present invention generally relates to the Internet wirelessly.
A method and system for accessing
Specifically, support multiple types of traffic
Wireless internet access system with capabilities
System (wireless Internet access system-WIAS)
You. [0003]Description of related technology Some wireless data systems have already been deployed
ing. In the early 1990s, the IEEE 802.11 standard was 902-9
28 MHz and 2400-2483.5 MHz industrial, scientific and medical (Ind
ustrial, Scientific, and Medical-ISM) for bands
Direct sequence and frequency hopping spread spectrum
 (Direct Sequence and Frequency Hopping Spread Spe
1ct and 2Mbps wireless devices using ctrum)
Stipulates interoperability standards and interfaces
Was. These systems are based on WaveLAN ™ (N
(Transferred from Luc Company to Lucent Technologies)
And embodied in products such as Proxim ™
Was. These IEEE 802.11 compliant systems are
Used for off-premise point-to-point applications
Have been. Other systems such as Metricom Ricochet (TM)
System is a mobile computing device that moves
Supports data rates up to 28,800 bps for
Was. [0004] Around 1996, digital cellular and
-Personal communication service (personal commu
nications service-PCS) system appeared. this
These systems range from 9.6 kbps (for IS-95 CDMA) to 32
 Supports data rates up to kbps (for PACS)
Was possible. In 1997, Unlicenced Natio
nal Information Infrastructure (U-NII) band
This band has three 100 MHz segments in the 5 GHz band.
(5150-5250, 5250-5350, and 5725-5825 MHz)
And made it possible to increase the data rate. Same as this
Sometimes complex signal processing machines
With multiple antennas
New systems emerge and wireless LAN applications
Data rates on complex multipath channels
Speed up. Such a system
One of which was developed by the European community
HIPE supports data rates up to 24Mbps
There is RLAN (trademark). RadioLAN (commercial
This system is compatible with wireless Ethernet.
Supports data rates up to 100 Mbps as an alternative system
Port. However, these and other conventional systems
Is a LAN application and point-to-point app.
Data rates previously reserved for applications
PCS type system mobility
 And can cover a large area
I can't. Therefore, these and other shortcomings of the prior art
For wireless Internet access that eliminates points
There is a need for a method and system. [0006]Summary of the Invention The system according to the invention comprises a mobile subscriber unit (m
obile subscriber unit) and existing
Send and receive multiple types of traffic between networks
Supports you. Multiple types of traffic
Support traffic transmission, and
A data structure that can adapt to the data volume of a constant traffic type
Structure is used. In addition, data transmission is a symbol
Inter-symbol interfer
nce) to prevent orthogonal frequency division multiplexing (o
rthogonal frequency division
ion multiplexing) and differential phase shift
Keying method (differential phas)
e shift keying)
You. Mobile subscriber unit and receiving unit in network
Knit is an antenna array that enables multipath transmission
(antenna array). [0007] A plurality of subcarriers
Symbols modulated above, each symbol being one
Or multiple data bits, where each of the bits is
Symbols whose combination represents a unique phase
The system according to the present invention for decoding
From superposition of modulated subcarrier
Includes means for receiving generated waveforms from multiple antennas
Where each modulated sub-carrier has a different frequency,
One of the serial symbols is assigned to the adjacent symbol of each pair.
Of the sub-carriers based on the phase difference between
It is formed by modulating on one. Book
The system of the invention further provides for each of the individual sub-carriers.
Means for extracting the response of each antenna
A vector with multiple elements equal to the number of
A means for forming for each carrier, where
Each element of the carrier vector is assigned to that particular subcarrier.
Represents one extracted response from multiple antennas
Means and number of vectors corresponding to adjacent subcarriers in each pair
Phase difference between adjacent subcarriers in each pair
And a mathematical set of adjacent vectors for each pair
The value of each symbol is calculated based on the phase difference obtained from the combination.
Means for determining. The above-mentioned outline description and the detailed description below
Again, there are only examples and explanations.
The examples and explanations do not limit the invention described in the claims.
No. [0009]DETAILED DESCRIPTION OF THE INVENTION Hereinafter, a preferred embodiment of the present invention will be described.
These examples are illustrated in the accompanying drawings and are described below in the context of the present invention.
Clarify in the light. In the drawings, in each drawing
Identical or similar elements have the same reference
Number. [0010] WIAS system architecture according to the present invention
Are key network functional elements, wireless interfaces
Source and critical radio ports and subscriber unit subsystems.
Consists of a stem. According to the WIAS system,
Large 17.5 Mbps broadband data pipe (broadband
 data pipe) to 32 kbps ADPCM audio.
Efficiently support a wide range of traffic types
Can be WIAS system, voice and real-time
Delay-tolerant (delay-t
olerant) supports low-latency data.
It is designed to be flexible to port. In addition, WI
AS systems are not only resistant to multipath
Harm (interference) and unlicensed bands
It has strong durability for reusing the spectrum inside. FIG. 1 shows a wireless interface according to the present invention.
Block showing Net Access System (WIAS) 100
FIG. WIAS100 is a 100 / 10BaseT Ethernet network
Network, asynchronous transfer mode (ATM) network,
Fiber network and token ring network
Design using an existing network, such as a network
Is possible. WIAS100 uses other networks
It is possible to use a design that uses Ethernet.
This is the preferred implementation, centered on the implementation for
Will be explained. As shown in FIG. 1, one WIAS 100 is provided.
Contains two or more islands 125, each eye
A land 125 may have one or more subnets 120
Contains. Subnet 120 is 100 / 10BaseT Ethe
Existing network (not shown) such as rnet LAN
Multiple wireless ports connected to the network (radio po
rt-RP) 110. Each subnet 120
At least one of the RPs 110 in FIG.
Has a mobility agent (MA) function as shown in
I have. More specifically, the RP 110 includes a plurality of RPs.
No connection with subscriber unit (SU) 105
Establish a wireless or wireless link and
It also manages terminations and changes. Each S
U105 is permanent or home internet
Protocol (Internet Protocol-IP) Address (H.IP)
 To the existing subnet 120 by assigning
Can be the home of Alternatively, Mobili
The tee agent, as indicated by MA 117 in FIG.
Virtual (virtual) independent of any wireless port
Can exist as a virtual subnet
Noh. In a virtual subnet, MA 117
Subnet (real subnet), or mobility age
A service that has multiple subscriber units in addition to the
SU 105 home agent not assigned to bnet
Acts as a Gent (home agent-HA). SU10
5, the function and structure of RP110 and MA115 and 117
This will be described in detail below. In addition to the subnet 120, each island
Reference numeral 125 denotes information transmitted to and received from the subnet 120.
Contains elements for routing (route selection) and processing
I have. These elements include router 130, network
Network management and monitoring system (network
management and monitoring system-NMM) 135
Gateway 140 and telephone gateway 145
It is included. The router 130 is connected to the network
A well-known device that routes information sent and received by
Yes, but it may be an existing part of the network.
You. NMM135 is the operation and management of WIAS100
And maintenance is performed as a single entity.
The gateway 140 communicates with the existing network and external communication.
Function of the interface with the communication system
That the network is protected by a firewall.
And many. The telephone gateway 145 is a telephone communication device.
Mobile identification of SU105 with the ability to support
Number (mobile identification number-MIN)
Map Internet Protocol (IP) addresses
(Mapping) function. In addition,
Gateway 145 is used to activate the voice IP modem for ongoing calls.
Sometimes. The data from the gateway 140 is
External communication such as the Internet
Transmission system, but the telephone gateway
Data from Way 145 is public switched telephone network
(public switched telephonenetwork-PSTN) 150
Can be transmitted. Data is stored at another telephone gate
Internet 160 and PSTN 15 via way 155
It is also possible to transmit between zeros. As shown in FIG.
In addition, the standard telephone 165 can be connected to PSTN 150.
Yes, IP phone 170 connects to Internet 160
be able to. Mobility function of subscriber unit One of the functions of WIAS100 is to support the mobility of SU105.
Is to port. WIAS100 has two levels of mode.
Supports mobility and only mobility within a subset
Outside, sometimes on different islands
Roaming between foreign subsets (roam
ing) is also possible. These two levels of mobility
Is a first-level intra-subnet with minimal equipment
Scaled for mobility
You. When wider area mobility is required
, Additional functional units are added. Mobility machine
Features include registration, location, and handoff
There is ability. The mobility within the subnet 120,
The first level where intra-subnet mobility is performed
Mobility is managed entirely by the RP110
You. The advantage of this method is that the single subnet 120
Small office using a LAN consisting of existing 100/1
0 Just connect RP110 to BaseT LAN to use WIAS100.
It can be used. With the expansion of WIAS100, additional
With the addition of mobility features, inter-subnet
Can support roaming between (inter-subnet)
it can. When a certain SU is activated, the SU is activated.
Will scan all available channels within WIAS100.
Search for the RP with the strongest signal available
You. In some cases, especially asynchronous unlicensed personals
Communication system (asynchronous unlicence
d personal communication system-A-UPCS) 1910-192
In the 0MHz band, multiple RPs can share a carrier
This indicates that the candidate carrier measurement algorithm in the SU is
Is considered by the This algorithm is based on the signal received from each RP.
Measure the received signal strength (RSS). S
When U is not transmitting or receiving, SU is
Tune into a band. If an RP is detected, the SU
Store the RSS associated with the issued RP in a table. Each
Table entry is valid for a short time (for example, 2 seconds)
Can be kept. SU is activated for the first time
Or the signal strength from the current RP is below the threshold
SU drops to the RP with the highest RSS in the table
Tune to a specific frequency and establish communication with that RP
Try that. Specifically, the maximum available RP signal
If found, the SU will send a random access slot (RAS)
 Sending a local registration message on the channel
Will attempt to connect to that RP, but
This will be described in detail below. The registration message indicates that the SU
Notify RP that you are trying to connect to the network
You. As a result of the registration, the RP has the SU's IP (SU.IP) address.
Address resolution protocol (address resoluti
on protocol-ARP)
Provide Upon receiving registration, the RP will receive a free ARP blow
Broadcast message (gratuitous ARP broadcast mes
sage) to the entire subnet and place it on the subnet.
All other hosts that are configured
You can update your ARP cache to reflect your
To do. As shown in FIG. 2, the SU transmits the radio
Register to port RP1. Moved to a new location
During transmission / reception of packets or SU
At periodic intervals when
Upon losing contact with RP1, the SU will return a candidate carrier measurement procedure.
Repeat the search for a better RP link. S
SU finds a new RP (RP2) as shown in Step 2
Then, a local registration message is sent out on the RAS. Ki
If the cached ARP entry already exists in RP2,
Enter the IP address (SU.IP) of the SU whose entry you want to register
RP2 is linked to the physical address of P1 (RP1.PA)
Send free ARP and everything else put on the subnet
The IP address of the SU to the physical address of the RP2 for the host
To map. When RP1 receives this ARP, it
Expire registration and stop proxy ARP response to SU
You. In the second level of mobility, the SU
Between subnets in a land or on different islands
Roaming seamlessly between sites
You. Support this inter-subnet mobility
Each subnet is a mobility agent (MA)
 Contains. SU is homed to the real subnet (beh
omed to) when the MA connected to the RP
Acts as a home agent and visits every SU
For the foreign agent (foreign agent -FA)
Work. SU to virtual (virtual) subnet
MA is acting as an external agent when
But there is no SU homed to the virtual subnet
Can act as a home agent
You. When intra-subnet traffic is large
Therefore, when large subnets are used,
It is preferable to use a SU homed to Bunnet.
But reduce the complexity of the RP used in the subnet
To do this, use a SU homed to the virtual subnet.
May be preferred. Furthermore, virtual
Prefer to use SUs homed to subnet
As a case, if many small subnets are used
And roaming from one subnet to the next
When a lot of time is spent on
Most of the network traffic is contained within the subnet.
Sometimes not. [0024] The home subnet of the SU is the permanent management of the SU.
Has a mask corresponding to the client IP address (H.IP)
Subnet. The external subnet is the SU
You can find yourself here on other subnets
And These two homing approaches (homing a
pproach) use one or both as needed
be able to. In either case, in the external subnet
Roaming SU (this is always the case
External subnets are used).
The care-of address (Care-o
f-Address-COA.IP). SU is Romin
When you leave the home subset,
The home subnet's MA) is the incoming packet addressed to that SU
If there is, trap the packet. Trapped
Incoming packets, with or without encapsulation, are COA.IP
Will be forwarded to the external subnet using. According to the encapsulation, the entire IP packet is
May contain sender and recipient IP addresses
Has a different IP packet with the new sender and recipient addresses.
Is included as the payload of the packet. Mobile host
Temporarily to another IP subnet with a known forwarding address
When a packet is relocated, the mobile
If the call arrives at the home address of the
Packets of mobile hosts in its new subset
Set the temporary address as "Recipient" and the home address as
It can be encapsulated as a "sender". Mosquito
When a packet is put into a new subset,
The packet is forwarded to the mobile host (routine
G), where the capsule was removed and the original message
Is recovered. The WIAS 100 has a corresponding host (correspo
nding host-CH) is appropriate, that is,
Node that sends packet to roaming SU
As a premise, full route optimization (full route optimizat
ion). Use route optimization
And the first IP packet sent from the CH to the roaming SU
Is forwarded via HA. The channel supports route optimization.
If it has, subsequent packets will be
Sent directly to the SU located at the care-of address COA.IP.
When the SU moves from one subnet to the next
Means that when an SU is registered in its new subset, a new CO
A. The IP address is acquired, and not only the SU HA
Binding with any CH with optimization capability
Is updated. Intra-subnet roaming (intra-
The registration procedure for subnet roaming)
Accept internet roaming
Has been slightly modified to In the external subnet
Entrants always register using their H.IP addresses.
You. RP is not home to a subset of that RP
RP receives the registration message from the new SU, the RP
Send a mobility registration request to the MA in the unit. MA, SU
The SU requires a message requesting authentication.
Send to home agent and use by HA for transfer
Give a care-of address (COA.IP). MA gives CO to SU.
Assign an IP address, and this COA.IP address
Used by the SU while staying in the department subnet. Another function provided by WIAS 100
Is fully seamless for all traffic types.
Support full seamless handoff
Is Rukoto. In a PCS system, the subscriber has a single voice
It is typical to have only a channel, but WIAS1
The link at 00 differs from this for a number of connections.
Connection-oriented traffic streams
Can be included. WIAS100 Handoff Pro
Settlement negotiates radio resources on new radio port
Low latency channels are set up preferentially.
Error-sensitive data is sent in its entirety
Is guaranteed. FIG. 3 shows a handoff process according to the invention.
It is a flowchart which shows. First, the SU will register for a new RP.
The negotiation is successful (step 305). Registration message
Is the IP of the RP (if any) to which the SU was previously linked
Contains the address. As shown in Fig. 2, the new RP (RP2)
 Sends a handoff request notification to the 100 / 10BaseT network.
Send to the old RP (RP1) via the work. RP1 responds with
And detailed TCP / UDP (transmission contro
l protocol / user datagram protocol-transmission control protocol
(Correlation / user datagram protocol)
Channel mapping (logical channel mapping-TLM) information
Messages containing information about the incompletely sent Dow
Link data back to RP2 along with the rest
(Step 315). RP1 is everything related to SU
Is discarded (step 32).
0), RP2 checks the available radio resources and checks each TCP
Establish a logical channel for each port connection
(Step 325). Constant bit rate (cons
tant bit rate-CBR) voice traffic first
Delay tolerant packet data is the last
As described above, based on channel priority, RP2
Assign a new logical channel to the connection
(Step 330). Logical channels that cannot be assigned
Some connections will have TCP timeout settings, if any.
Failure may occur based on the fixed value (step 33
5). But all successful logical channels remain
Is left. Connectors dropped during handoff
To minimize the number of applications (TCP / UDP ports)
The system ensures that all RPs are
System traffic and downstream traffic
Can be realized to support. For example, each
Wireless ports always support certain voice and packet data functions.
It is possible to realize to support. is there
If you always support certain voice features,
Port to support incoming voice connections.
It is not necessary to rearrange the frame structure. As well
In addition, the packet access mechanism in WIAS100 is dedicated
Although not optimal for data streams, this mechanism
Until the appropriate logical channel can be established.
Lower the quality of service (QoS)
Temporary support for TCP / IP data connections.
It is possible. The wireless port realized as above is good
Even better, the traffic supported by WIAS100
Depending on the type of rack, a more suitable frame
There are other structures. Voice and real-time multimedia
support Among the supported traffic types, WIAS1
00 is especially useful for voice and other real-time multimedia
Traffic can be handled. VOIP (voice-
over-IP: handles voice / traffic using IP
In order to do so, WIAS 100 is an H.323 protocol framework
Can be used. This protocol framework
Is an ITU-T H.323 International Telecommunications Union
For local area networks that provide network quality
Visual Telephone System and Equipment
Equipment for Local Area Networks) recommendation (H.323 IT
U Telecommunication Standardization Committee, Geneva, Switzerland, May 199
6), this recommendation does not apply to packet-switched networks.
Real-time, multi-point, multi-
For video conferencing
And supports IP-to-PSTN networking
You. According to the present invention, as shown in FIG.
To support traffic with WIAS100,
There are at least two broad categories. First mosquito
Tegori is a multimedia cordless phone scenario
How to have a sound with direct connection to PSTN150
As shown by the voice network 175, the individual voice interface
Ace. In this case, the audio
Is Internet connectivity
Not required, voice and data are treated separately. Mobi
Is 100% BaseT data network, even in voice networks.
Network or both
is there. Voice traffic supported by WIAS100
Another category is to transfer voice and data to WIAS100
Integrated on top. Voice and data are integrated on WIAS100
There are many scenarios to be combined, some of which are shown.
Then, it is as follows. (a) Internal telephony gateway
(internal telephony gateway-TGW) Via 145
From SU105 to PSTN telephone 165 (PSTN-Phone)
connection. This gateway is an IP-based WIAS10
Function of network interface between PSTN0 and PSTN150
And pass-t on the voice network 175 (pass-t
hrough) connection. (b) PSTN connectivity
SU 105 via Internet 160, not required
To IP telephone 170 (IP-Phone).
(c) S via Internet 160 and external TGW 155
Connection from U105 to PSTN-phone 165 and
And (d) the Internet 160 and the PSTN 150
External TGW tying. [0034] These four voice connection scenarios
Is shown in FIG. These voice connection scenarios
In order to realize Rio, WIAS100 is an IP-based real
Time protocol (RTP) can be used. What
About this protocol, H. Schulzrinnet et al.
"RTP: transport for real-time applications
Protocol (RTP: A Transport Protocol for Real-
Time Applications) ”(Request for Comments
Associate) RFC 1889, Internet Engineering Task Force, Dec
ember 1997), but this is user data
User Datagram Protocol (UDP)
Handling voice traffic using connection
You. RTP supports timing reconstruction, loss detection, and security.
Prototyping and support for content identification
It is a col (thin protocol). H. Schulzrinnet et al. `` R
TF: transport for real-time applications
Protocol (RTP: A Transport Protocol for Real-Time
 Applications) ”(Request for Comments (proposed standard)
RFC 1889, Internet Engineering Task Force, Decembe
r 1997) IP-based RTP control
Protocol (RTCP) for monitoring RTP sessions
It developed QoS feedback and returned RTP
Can communicate information about who is participating in the
This is especially true for subscribers who have ongoing multimedia sessions.
This is useful when participating in an event. RTP and RTCP
Both are implemented in the application layer of the H.323 protocol
It is possible to do. Voice in Voice / Data Integrated Category
In two of the connection scenarios, dual tone multiplexing
Frequency (Dual-Tone Multi-Frequency-DTMF)
Must be transmitted over an IP network. DTMF
Gunnaring is a VoIP infrastructure that uses RTP.
Supported on the key. H.323, on the other hand,
I running over TCP for connection and control
H.245 connection control protocol defined by TU-T
Is supported. Therefore, H.245 is an alternative to RTP
DTMF signaling can be supported. Radio Interface and Logical Channel Structure WIAS100 supports dynamic time division duplex / time division multiple access (d
dynamic time divisionduplex / time division multiple
access-D-TDD / TDMA)
it can. This type of system has an aperiodic variable frame
Using structure. FIG. 6 shows a WIAS 100 according to the present invention.
2 shows a frame structure used in the first embodiment. As shown in FIG.
As such, each of the data transmissions is divided into a series of phases,
Phases are for a specific time interval, for example, 30 seconds
Lasting. Each phase has multiple superframes (superf
rame). 250 ms duration (mse
In the superframe of c), the 30-second phase has 120 superframes.
It will include Pafram. Each superframe is an example
For example, it consists of 64 frames, which is 250msec
Superframe is equivalent to about 3.9 msec per frame.
Will be. Each frame has multiple time slots (eg,
For example, 16), including the uplink slot
And both downlink slots are included. Frey
Uplink slots and downlinks used in the system
The number of slots is determined by one superframe and the next superframe.
Can vary between frames. Each frame starts each frame.
Downlink node control channel (no
de control channel-NCC)
No. The NCC has defined a phase frame format.
It includes the length and configuration of each frame.
You. In addition, all uplink and downlink ties
The time slot is defined with respect to the NCC immediately preceding it.
Have been. The NCC contains paging information and this information
The information includes the incoming connections directed to a particular SU,
Notifies SU that there is a channel or slot assignment
Used to In addition, NCC
Lot confirmation and word error notifications (acknow
ledgement and word error indications-ACKS / WEI)
Also do. At the beginning of each superframe, the NCC
Whether the logical channel of the type is supported by the RP
Frame description to notify the SU along with the associated time slot
Contains a frame descriptor. NCC is free
Only one per system. Each frame ends each frame.
Uplink access slot (rand
om access slot- RAS). R
The AS is divided into several frequency sub-channels, each
The subchannel is composed of 32 subcarriers. First sa
Channel block is downlink acknowledgment (ACK / NAK)
 Reserved for use. Other blocks are access messages
Reserved for registration and registration messages. Access
The message indicates that the SU is UDC (uplink data channel-uplink).
Link data channel), UVSC (uplink video stream ch
annel-uplink video stream channel)
Or UMVC (uplink multiplexed video channel
Link multiplexed video channel).
This is described below. Registration message at RP
SU to notify you that you are connected to the network
Used by Access message and registration message
Pages are randomly selected to reduce collisions.
Of 32 subcarriers, limited to the selected block
Sent using the set. RAS is an uplink
By the subscriber unit to generate an imslot request.
Even used. Between the NCC and RAS, a series of time slots
Can propagate various information as shown in Table 1 below.
Can be. The frame should preferably be
Link time slot (RP to SU), followed by
Included uplink time slots. Each time
The slot is separated from the preceding NCC by time.
Is defined by [0040] [Table 1-1][0041] [Table 1-2]What type of traffic is supported by the RP
RP is DDC, D
Any combination of PSC, DVSC and DMVC channels
You can choose and support them. Uplin
RP uses UDC, UPAC, UVSC, and UMVC channels
Can support. As mentioned above, each wireless
The card preferably supports voice and packet data functions.
It has been realized to be. Therefore, the typical frame
Configuration, at least one DPSC / UPAC pair and one
DMVC / UMVC pairs are used. The remaining one in the frame
0 time slots allocated to DDC and UDC channels
It is possible to RP is a multiplexed voice slot, packet channel
The frame configuration including the channel etc. once every 30 seconds phase
Only can be changed freely. DDC / UDC channel
From a certain total, the RP is assigned to downlink and uplink.
The assigned fraction is taken for each superframe (250 msec)
Can be changed freely only once. RP
DVSC / UVSC ratio can be changed at the start of paframing
it can. RP is DDC / UPC ratio, DVSC / UVSC ratio or frame
When changing the system configuration, the RP
Make sure everything moves seamlessly
To prevent the logical channel being used from being deallocated.
Have a responsibility to Spectrum management Spectrum management at WIAS 100 has several dominant sources.
It is based on reason. First, RP is generated on the same channel
Sends co-channel interference only when needed
Minimize by doing trust. Second, RP
Frequency channel only when the traffic load is low.
Has weakened control over Le. Third, RP is a dynamic frequency
Incorporate quotas for different traffic patterns and
Frequency allocation to accommodate different types of interference
You can change the guess. RP frequency selection function
Being autonomous. Fourth, the RP has a fixed periodic
By notifying that there is an RP in the Tarval,
Random access and sleep on uplink
 Support mode. Fifth, RP is another non-
Camp-on the active RP frequency,
The channels can be substantially shared. Sixth
In addition, the two RPs with low activity have a sleep mode function.
Sharing frequency channels by sacrifice
Can be. Finally, when the SU is in the inactive state
SU predicts that RPs will exist over long intervals.
I don't think In other words, RP with very low load
If the SU is idle in the call corresponding to
Wake up from sleep cycle in sync with selected RP
Will be. But when the SU wakes up, the RP is a channel
When the SU learns that he has left, he places the new candidate RP in the correct order.
To start looking for (same RP on another channel
You may find them). To obtain a channel of operation,
Each RP scans the channels in the band of operation.
To obtain the carrier frequency. Uh
And the RP should have minimal interference or power
Operational checks where the bell is below a predetermined threshold
Select a channel. RP includes low, medium, or high U-NI
Operates on the I band or a combination of these bands
Can be instructed to do so. 5.1 for U-NII band
50 GHz to 5.350 GHz and 5.725 GHz to 5.825 GHz
Spectrum, and this spectrum is 300 MHz
It corresponds to the entire spectrum. WIAS100 can be realized in other bands
Thus, improvements can be made. For example, Listen
-By complying with Before-Talk rules, WIAS100
Used in the 1910-1920 MHz asynchronous unlicensed PCS band
It is possible to realize it. Another way
And WIAS100 is licensed PCS band (1850-1910
And private spectrum including 1930-1990 MHz)
It is also possible to use. RP is preferably each candidate
Monitor channel and at least one superframe
And then select an operation channel
Like that. A channel within a certain band is scanned, and
The process of selecting the complement channel depends on what bandwidth the RP
It depends on whether it is working. RP 2.6 MHz van
RP, within each 100 MHz bandwidth,
Start searching for a minimum of 18 MHz frequency and
If the channel is not at least 18 MHz, search outside that area
Only try to do. Channels are 1.5 MHz apart
Channel number 1 is 1.5 MHz from the bottom of the band.
Begins with z. Channels 1-11 use the initial search set
Make up. The total search set must include the entire band
And the entire band has 66 channels (1 ... 66)
I have. When the RP is operating in a 5.2 MHz bandwidth
The RP is 18 MHz and 51 MHz from the bottom edge of the band
Start looking for channels between. RP once
Considering two channels, the search interval is 3 MHz.
You. The initial search set has 10 possible channels
, Including channels 14, 16, 18, ..., 32
I have. RP finds suitable channel in initial search set
If not, the RP searches in other channels.
The total channel set is 32 (2, 4, 8,
 ..., 62, 64). When RP is operating in 10.3 MHz bandwidth
The RP is 51 MHz and 100 MHz from the bottom edge of the band
Start looking for channels between. RP at once
Considering two channels, the search interval is 6 MHz
You. Initial search set is 7 channels 38, 42, 46, 50, 5
Includes 4, 58, 62. RP becomes one of these bands
If no channels are found, the entire 16 channel set
Body (2, 6, 10, ..., 58,62)
You. Finally, the RP operates at 20.6 MHz bandwidth
RP is 51 MHz and 10 MHz from the bottom edge of the band
Start looking for channels between 0 MHz. RP
Considering 8 channels at once, the search interval is 12 MHz.
ing. Initial search set for 3 channels 42, 50 and 58
Contains. RP finds channels in these bands
If not, the RP will use the entire 7 channel set (10, 18,
26, 34, 42, 50, 58). Reduces battery life and minimizes disturbance
WIAS100 has a built-in power control
ing. Power control is both open and closed loop
This is done using one of the methods. In the open-loop approach, the SU is
Set the initial transmit power based on the transmit downlink power.
So that sufficient signal strength is obtained at the RP receiver.
I do. In a closed-loop approach, the RP needs its power level
Can be instructed to increase or decrease according to. To support sleep mode, NC
C is the superfuffer at least once per superframe.
Reappears at the beginning of the frame. The NCC is more frequent
Pages can be sent freely, but the NCC
Sending the SU
The purge of any unresolved pages, if any,
Must be transmitted starting from the beginning of the frame. SU
Will display all outstanding pages, regardless of their sleep state.
Until the NCC indicates that the flash has flashed.
The first NCC of the frame needs to be monitored along with subsequent NCCs
There is. WIAS 100 is an autonomous frequency selection by each RP
NMM will either let go of the channel or
Or instruct RP to acquire a specific frequency channel
Have the ability to All active radio ports
Is superframe irrespective of traffic level
And once to send an NCC burst. By this, the nearest
SUs have access to the uplink channel
ing. If there is no traffic for some phase at the RP, the RP
Marks itself as idol. Idol RP
Transmit one NCC burst per superframe
Continue. SU enters RP without becoming active
Can be recorded. NCC sent from idle RP
Contains a marker indicating that the RP is idle.
You. If the monitoring RP sets the appropriate frequency channel elsewhere
If you can't find the RP,
The same carrier used by the active RP
NCC transmission can be started. RP being monitored
Or when one of the new RPs becomes active,
The active RP uses the Ethernet network to
Send a message to the RP and ask the channel to abandon.
Request. The idle RP is periodically (every superframe
1 burst) to transmit, other idle RP and channel
Each RP that initiates a NCC collision on the downlink
Adjust its superframe timing to avoid
There is a need to. The two active RPs have different frequency
It is also possible to share bands. Spectrum is pre-
When in Miam, the network goes into sleep mode
Prohibit and can operate in aperiodic mode. This model
Mode, once (or less) per superframe.
Above) Instead of sending, each RP is 250 from the beginning of the leading NCC
The NCC must be sent within msec. Either
Each RP takes 200 microseconds before sending any frame
Monitor the channel during (μs). Channel in use
Otherwise, the frame starts with the NCC first.
At the end of the frame, the RP waits at least 1 ms
A retransmission must be attempted. Send attempt fails
And CSMA / CA method with a randomly increasing back-off period
Will be imposed. Until the RP sends the required NCC
After 250 msec, the RP finds another frequency.
To work there. RP cannot acquire channel during extension period
NMM will shut down under control as needed.
You can manually request the RP to perform an action.
After that, the RP should automatically obtain the frequency channel
Also assign specific channels according to NMM instructions.
Both are possible. Physical radio channel WIAS 100 preferably employs orthogonal frequency division multiplexing (Orth
ogonal Frequency Division Multiplexing-OFDM)
Information is transmitted between the SU and the RP using an expression,
Uses a very large number of simultaneous subcarriers,
Each is modulated using a low symbol rate. FIG. 7A
And FIG. 7B show OFDM transmitters according to the invention, respectively.
FIG. 2 is a block diagram showing a transmitter 200 and an OFDM receiver 250.
You. High data rates with a single subcarrier in OFDM systems
Instead of using a very low symbol rate.
Used for a number of subcarriers. Symbol level for each subcarrier
The rate of inter-symbol interference
ol interference-ISI)
ing. FIG. 8 shows the OF of FIGS. 7A and 7B according to the present invention.
FIG. 4 is a flowchart showing the operation of a DM transceiver / receiver.
You. First, N_cN symbols_cModulated on subcarriers
Inverse fast Fourier transform
 transform-IFFT) is used (step 80)
5). The resulting time waveform is N_cWeights of the modulated subcarriers
Tatami, so the channel is upconverted
Transmitted above (step 810). On the receiver side,
Fast Fourier transform to separate each individual subcarrier
 (fast Fourier transform-FFT) is used (step
815). After the subcarrier is separated from the transmission signal,
The response of each antenna included in the antenna array is
Extracted (step 820). Next, the extracted les
A vector is generated for each subcarrier based on the response
(Step 825). Each vector is
Contains a number of elements equal to the number. Specific sub-transport
The elements included in the wave vector are the
Represents the extraction response of the receiving antenna to the waves.
You. Using these vectors, the neighbors of each bear
The phase difference between adjacent subcarriers is related to the vector of adjacent subcarriers.
To perform mathematical operations such as dot product
(Step 830). Each pair
When the phase difference between adjacent subcarriers is determined, the subcarrier
The symbols used to phase modulate each of the
(Step 835). These symbol
Phase encoding the subcarrier
And then decoding them.
This will be described in detail below. The symbol rate applied to each subcarrier is
Because it is small, there is time dispersion in the channel between RP and SU
Nor does ISI occur. Caused by the channel
Frequency spread due to transmitter and receiver defects
As long as is small enough, the individual subcarriers are almost
Since they remain orthogonal, independent reception
it can. The following describes the OFDM transmission according to the present invention.
Of the operation of the OFDM receiver 200 and OFDM receiver 250
I will tell. The transmitted OFDM signal, that is, transmitted between the RP and the SU
Baseband complex envelope of the signal being received (baseba
nd complex envelope) has the form
You. [0063] (Equation 1) In the above, r (t) is applied to the transmitted waveform.
Pulse shape. With a typical uplink signal
Implies that each burst contains a single OFDM symbol period
It is assumed that In other words, each sub-carrier is simply
After transmitting one symbol, the transmitter
Do not transmit for a certain interval before the strike. Equation 1
7A, the sum portion indicated by sum 208 is:
Subcarrier and symbol Z_iWith the IFFT 206 of FIG.
The result added to the mapping 204 is shown. Uh
And this sum is changed by r (t), which is
It is represented by The resulting waveform is then
Analog (digital-to-analog-D / A) converter 21
2 is converted to an analog signal by U / C 214
Up-converted and transmitted by antenna 216. Symbol Z representing data to be transmitted_i
Is the frequency domain difference quadrature phase shift keying (fr
equency domain differential quadrature phase shift
 keying (FD-DQPSK) and other differential encoding (differentia
l encoding) used to modulate each subcarrier.
Used. Symbol Z_iHas the following format:
You. [0066] (Equation 2) Φ_iHas the following format: [0068] (Equation 3) Where n = 0... M−1 and M is a difference sign
The number of phase levels used in the decoding. Therefore, FD-D
In QPSK, M = 4 and the different phase levels are Φ_i ε {π /
4, 3π / 4, -π / 4, -3π / 4}. Next, the video of each pair
Is the phase difference between two adjacent carriers (Φ_i -Φ_i-1) As sign
Be converted to Minimum spacing between two carriers is 1 / T_sEqual to here
In T_sIs used to sample the signal on the receiver side.
The length of the time window used. The phase modulation signal is transmitted on the channel between the SU and the RP.
When transmitting, the SU and RP transmit using multiple antennas
A signal can be received. But multipathing
Channels may cause distortion of the received signal.
You. The SU and RP at WIAS 100 are preferably as shown in Figure 7B.
The plurality of antennas 252 of the
Because it uses a receiving antenna, it changes over time,
For multipath channels that spread intermittently,
Tor models are available. [0071] (Equation 4) Each element of the vector h (t, λ) is
Channel between transmitter and one element of receiver array
Is represented. This model is received from the transmitter
N to ba_LIndividual passes, each of which
The receiving array is in the form of a single plane wave. Each component
Event-direction-of-arrival-DO
A) θ₁, Scalar path gain α
₁, The path delay τ relative to the first arrival path₁, Doppler
 (Doppler) shift ρ₁, And an additional constant phase coefficient γ₁
Is characterized by: Vector a (θ₁) Is the direction
θ₁Response of the array to a plane wave incident from
Representing steering vector
It is. All of the delays are based on the first arriving component.
Is τ₀= 0 is normalized. In the array
Can be expressed as follows. [0074] (Equation 5) In the above, each element of u (t) is
Received by one of the antennas 252 in the antenna array
Represents the signal obtained. Here, the pulse waveform is
Very gradual change in response to chippath component delays
The received signal is as follows:
Can be written in [0076] (Equation 6) As shown in FIG. 7B, each of the antennas 252
The signal received by each is down-converted by D / C 254.
Is converted to a digital signal by the A / D converter 256.
It is converted and pulse shaped by the shaper 258. Sa
Furthermore, the signal received by each antenna is
Are separated into each subcarrier. Extract symbols from subcarrier
Delta function 262, multiplier 264, and
And v using adder 266_kIs made. Be
Kutor v_kIs the antenna of the antenna 252 with respect to the subcarrier k.
Shows Ray's response. [0078] (Equation 7) The vector v_kEach element is a subcarrier
one response of the antenna included in the array for k
Represents The subscriber amplifiers have non-zero lamp amplifiers.
The ramp-up and ramp-down times are related.
Linked, minimizing inter-subcarrier interference
R (t) is the square root raised cosine pulse (square
e root raised cosine pulse) (Fig. 7A
Written as time analog of pulse shaping represented by 210)
Can be [0080] (Equation 8) Where T_W= T_S+ 2ΔT. Therefore, [0082] (Equation 9) Ρ₁ << 1 / T_sDoppler shift is small like
When, V_kCan be expressed as follows. [0084] (Equation 10) The vector v_kAfter the adjacent sub-carrier is made
The phase difference between may be calculated using detector 270
it can. J. C. Liberti simplifies calculation
"Smart antennas are orthogonal frequency division multiplexed (OFDM)
Techniques for Applying Smart Ante
nnas to Orthogonal Frequency Division Multiplexing
(OFDM)) '', Bellcore Engineers' Notes, October 1997
, This approach is based on the continuous subcarrier v_kwhile
To form the inner product of
You. This technique multiplies continuous-time symbols by symbol
Time-domain difference demodulator to estimate the phase between
Similar. In FD-OFDM signals, the carrier-to-carrier
Phase shift multiplies the continuous carrier space vector together
It has been estimated by [0086] (Equation 11) Y representing the phase difference between adjacent subcarriers_kof
The expected value is calculated using the phase angle {γ_l} Over a set of
Is vector [Outside 1] If you take over, it will be as follows. [0088] (Equation 12) However, in practice, the phase of the individual carrier is
Work randomly for any particular OFDM burst
No fading (fading)
ing. Separately, assume that the DOA distribution is even
And Y_kThe expected value for all directions of arrival was as follows
I can take it. [0090] (Equation 13) In a linear evenly spaced array,
You. [0092] [Equation 14] When M = 2, 4, 6, and 8, when Δx = λ / 2,
Is Ψ (M) is 1.09, 1.17, 1.21, and 1.24, respectively.
Become. When M >> Ψ (M), the product term (cross prod
Even if uct term) is ignored, the same result as Equation 12 is obtained.
Therefore, the phase set and DOA are independent of which particular burst
Non-random, but constant phase interaction
And DOA distributions are both Y_kThe contribution of the product term to
Tends to minimize. Therefore, Y_kIs
Approximated. [0094] (Equation 15)However, [0096] (Equation 16) The information is obtained using the phase difference between successive symbols.
Y is encoded_kIs the modulation method decision variable (decision
 variable). When the approximation of Equation 15 is valid (that is,
(Using one of the two justifications given)
The approach is multi-pass resistant and multi-pass components
Delay τ₁ Differential rotation between the subcarriers caused by the
Caused by the residuals from the multipath channel
It is only strain. For example, if the subcarrier separation is 5 kHz
The maximum component delay (the first arrival component)
If the time is less than 5.5 μs
Since the phase of the exponential term at
The sum of multipath components is mostly real
Become. So Y_kIs approximated using the following equation: [0099] [Equation 17] When the phase of the sum term in Equation 15 is close to zero,
If so, Y_kIs biased even when the noise is negligible.
Is an estimated value. The detector of (10) is calculated using the following equation.
Slight improvements to account for sampling offsets
Have been. [0101] (Equation 18) According to this method, the received signal is
T₀Need to find the value of
The best decision metric. [0103] [Equation 19] The method described here uses WIAS100.
Used in the uplink. Dow at WIAS100
Link uses a similar technique, but with a circular prefix.
A prefix (cyclic prefix) is added to each burst.
This cyclic prefix is preceded by the transmitted data symbol
The corresponding symbol is added, and the resulting waveform is
Circular convolution of pulse response and transmitted data (ci
rcular convolution)
Therefore, the FFT processing on the receiver side is simplified. According to the described method, the obtained
The result is additive white gaussian noise.
ussian noise-AWGN) Optimal multi-sensor for channels
Within 2 dB of coupling and in multipath channels
Minimum Mean Square Error (MMSE)
 Outperforms single-tap beamforming
You. This method works when there are timing and frequency errors.
It also demonstrates excellent performance. However, the non-
Main features of non-coherent combiner
The main advantage is that training is unnecessary and burst data
Works effectively, eliminating the need for tracking between bursts
It is to be. So this is supported by WIAS100
Be suitable for all data types. How Differential Encoding and Decoding Are Performed
In order to explain, the 6 subcarrier differential QPSK OFDM system
Will be described as an example. The data to be sent is {S_k} = {1,0,
0,1,1,0,0,0,0,1}. In the first step,
The data is split into pairs of bits, which are then
Are mapped to the phase difference as shown in Table 2. [0107] [Table 2] The phase value {Φ_i} Is the modulation symbol Z_i = e^jΦi
Is used to determine the transmit waveform described in Equation 1.
Is done. On the receiver side, the phase value {Φ_i} Is the expression 11 to 18
As described above, the recovery is performed as follows. [0109] ｛Φ_i} = {Π / 4, -3π / 4, -π / 4,3π / 4,3π / 4, -3π / 4} Φ₀ = π / 4 and ΔΦ_i  = Φ_i -Φ_i-1That it is
If known, the set of differential phase shifts is {ΔΦ_i} =
{X, π, π / 2, π, 0, π / 2}, where "
X "is Φ₀Does not depend on any leading phases.
You. ΔΦ₁, ..., ΔΦ_FiveWhen used together with Table 2,
The constant reception bit is {S_{k (received)}} = {1,0,0,1,0,0,0,0,1}
 It turns out that it is. Noise that breaks the receiving channel
Transmitted bits are not received because there is no
I have been trusted and recovered without error. In differential coding, comparison is made with non-differential modulation.
There are several advantages. First, the data is between consecutive symbols
Since the whole is encoded as the difference, the carrier recovery method
It does not need to be used to estimate the absolute phase of each symbol.
In addition, the data is encoded by the phase difference between successive subcarriers.
The modulation technique is caused by the channel
Phase change. Due to channel
The phase change depends on the signal (N_cΔf) over the entire bandwidth.
Between the subcarrier and the next subcarrier.
Between them tends to be smaller. In addition, the differential encoding method
According to the efficient signal from the multi-antenna element
And almost optimally. [0111] Differential coding is not available for AWGN channels.
The performance is slightly lower than that of differential coding, but the difference is
Not so noticeable in the chippath channel. Used in the above example
In addition to the M = 4 modulation scheme used, the differential phase shift of M = 16
The keying method is generally used. Also, OFDM
In the system, the number of subcarriers N_cIs used in the above example
It can be much larger than six subcarriers.
Typical values range from 64 to 1024 subcarriers.
For example, if the operation channel is 5.15 GHz,
If the number of transmissions is 64 and separated at 5 KHz,
The transmission starts at 5.15 GHz on the first subcarrier and ends on the last subcarrier.
It will be in the range of 5.15 GHz + 315 KHz. Problems common to all OFDM systems
Is the peak-to-average power ratio
PAR), the power ratio of which is crest factor (cr
estfactor). For this, R.A.
 Ziegler, Peak Envelope in Multicarrier Modulation.
Reducing Peak Envelope Fluctuations
in Multicarrier Modulation), Bellcore Engineer's
 As described in Notes, February, 1995, its contents
Are included herein by reference. QPSK subcarrier variation
With modulation, the total energy of each burst is
It is constant regardless of the data. But Bath
The peak instantaneous power of the
If the average power of the burst is N_cMay be doubled.
Average power has energy per burst in bursts
Defined as divided by duration. N_cIs big
Can produce unacceptably high peak power values.
Very expensive, high dynamic range
If the transmission amplifier is not used,
It is the only cause. But this transmit amplification
Vessels are not power efficient. To solve this problem, each burst
Is the value of crest factor reduction (CFR)
N_pSubcarriers are reserved. N_pOptimal transport
PAR is a systematic algorithm for finding wave sets
Is not easy to realize. In addition, PA
A set of subcarriers that minimizes R, that is,
It is not strictly necessary to find only the subcarrier
No. Therefore, a random search technique is used. This
In the method, N_c -N_pN data subcarriers_p PAR
Used with carrier and N for split CFR subcarrier_p
Set of modulation symbols (X_i} Is used. OF
DM bursts contribute from data symbols and CFR sub-carriers
The sum of the contributions from the waves can be expressed as
You. [0114] (Equation 20) The data part of the modulated signal is transmitted in air.
Data symbol set {Z_i} Is calculated once for each
You. Set of random values {X_i} Is a QPSK symbol set
Drawn from the s_p(t) is calculated. Peak vs. average power
-Ratio is the resulting burst s_tCalculated for (t).
PAR is T_ppIf smaller, set of PAR reduction symbols
{X_i} Is received, s_t(t) is transmitted. PAR threshold
Value T_ppIf larger, the new {X_i} Set selected
And the process is repeated. {X_i} Set, search
Minimizes PAR across sets, so it is always kept
I have. T_ppYou can get a smaller PAR, {X_i} Suitable set
Is N_ppMinimum PAR if not found during repetitions
{X_i} Set is used. This feature is
PAR estimator 218 in 7A transmitter 200, PAR
The symbol 220, the IFFT 222 and the shifter 224
Is provided. In addition, the reverse of this function,
The stripping out is shown in FIG. 7B.
By the strip PAR symbol 268 in the
Done. Table 3 shows the desirable system specifications of WIAS100.
Are shown for different channel bandwidths. This
These system specifications have been shown only as preferred specifications.
So, even if it is limited to these, it cannot be changed
Not even. [0117] [Table 3]As shown in Table 3, all terminals
It does not necessarily support these data rates. Concrete
Typically, portable subscriber terminals are 2.6 MHz and 5.2 MHz
Only the width is supported. Typical example is low cost
Desktop systems can support up to 10.3 MHz
Links that require a large amount of bandwidth
Command width can be supported. Radio Function Architecture FIG. 9 is a block diagram illustrating the RP 110 according to the present invention.
You. RP100 is a POTS / ISDN interface module
900, Ethernet transceiver 905, IP interceptor
910, mapping multiple TCP / UDP ports to logical channels
Module (TUPLCM) 915, Mobility Management Module
920, security management module 925, logic
Channel Module (Logical Channel Module-LCM)
930, 935 and 940, logical channel multiplex
Remixing module 945, modulation / coding / crest factor low
Reduced symbol module (MECFR) 950, transmission
Antenna selector 955, OFDM baseband processing module
Modules 960, 964 and 968, RF transceiver
(RF XCVR) 970, 975 and 980, multi-antenna
Coupling module 985, detector / receiver 990,
And logical channel demultiplexing module 99
5 is provided. [0120] POTS / ISDN interface module 90
0 indicates that the input serial data stream is
Mapped to the ream 902 (association) and output data
Stream 902 is a modulated data that can be transmitted on a POTS line.
Data stream, or transmitted over an ISDN line.
It may include a digital data storm that can be sent
is there. Output data stream 902 is output to TUPLCM 915.
Received in the POTS mode or ISDN mode.
In both cases, be bi-directional
Can be. An Ethernet transceiver 905 is an Ethernet cable.
Data from the serial data stream 90
8 and this serial data stream is
Internet Protocol (IP) packets
Received by the ceptor 910. After that, IP in
The ceptor 910 converts the IP packet into a serial data stream.
And transfer it to the TUPLCM 915. Each TUPLCM 915 is associated with a subscriber
And the transferred serial data stream 908
Or one of the data streams 902 is received and
Traffic of individual LCMs 930, 935 and 940
Transfer to TUPLCM 915 is the LCM 930, 935 and
And 940 to different TCP or UDP port numbers
You. TUPLCM 915 is an object that uses software.
It can be realized as a project. Therefore, TUPL
CM 915 is the active currently associated with each RP
Create as needed to match the number of subscribers,
And can be deleted. Mobility management module 920 notices
(Care-Of) Map the address to the real IP address and
Ensure that Vail IP is supported. Security tube
Management module 925 performs authentication and accesses WIAS100.
Make sure the subscriber you are trying to access is authorized. Each LCM 930, 935 and 940 has a TUP
With a first bidirectional input port connected to the LCM 915
I have. This first input port is associated with a particular subscriber
Support data for specific TCP / UDP ports
are doing. The data flowing into the first input port is
Buffered for transmission from the gateway. Each LCM 9
30, 935 and 940 are logical channel multiplexes.
The second input port connected to the
Have. The data appearing at the second input port is the first input port.
Buffer for transmission from power port to TUPLCM 915
Is placed. Each LCM 930, 935 and 940 is the first input
The delay-tolerant transmission data from the
Automatic repeat request (ARQ) response
Is buffered until it is received from the subscriber. no
Negative acknowledge-NAK is received
And each LCM 930, 935 and 940 is buffered.
The transmitted data is retransmitted from the output port. Preferably
Is delay-sensitivity, such as voice
ive-delay-sensitive)
There may not be. As with TUPLCM 915, each LCM 93
0, 935 and 940 are objects that use software.
Each LCM 930, 935
And 940 support each subscriber's data stream
Can be easily created or deleted. The logical channel multiplexing module 945 includes:
Data from the output of each LCM 930, 935 and 940
Stream and receive the data stream as a single data stream.
Multiplexed on the stream 948, which is the MECFR 950
Received by Extracted from each data stream
The amount of data that can be
Matches the number of bits contained in The MECFR 950 is included in the data storm 948.
Group the set of bits
And generates a program 953 as an output. MECFR950 is the error
Forward Error Control to control
rol-FEC) The coded bits and the
Crest factor reduction to minimize peak-to-average ratio (Cres
t Factor Reduction-CFR) bits to data stream
948, required dynamic level of RP110
And the transmission efficiency is improved. MECFR9
50 groups the blocks of coded bits into one group,
Inverse Fast Fourier Transform (in
verse fast Fourier transform-IFFT)
Orthogonal Frequency Division Multip
lexing- OFDM) Create a symbol. After that, tran
The transmitter antenna selector 955 outputs the resulting data
The trim 953 is connected to the RF XCVR 970, 975 and 98
Map to one of 0. Each XCVR 970, 975 and 980 has a
Digital input, digital output, and antenna port
It is. Is each XCVR 970, 975 and 980 an input?
These digital data are converted to analog signals and converted to RF carriers.
And amplify the resulting signal and pass it through a filter.
And transmit from the antenna port. Receive from each antenna
Filtered data, amplified and down-converted
Are digitized and made available at the output
You. Each OFDM baseband processing module 96
0, 964 and 968 are one XCVR 970, 975 and
And 980, receiving digitized symbols from the outputs of
Extract timing information using Fast Fourier Transform (FFT)
And extract modulated symbols on each individual subcarrier
And generate an output data stream. The multi-antenna coupling module 985 includes:
Each OFDM baseband processing module 960, 964 and
And the output data stream generated by
Using the differential OFDM multi-antenna combining method described above.
To generate a combined data stream 988
You. The detector / receiver 990 is a data stream.
Symbol 988 in the binary digital data stream
And the wave height included in the data stream 988
Eliminates rate reduction (CFR) bits and error control coded bits
Thus, a data stream 993 is obtained. After that,
The physical channel demultiplexing module 995 is
Nari digital data stream 993 is received and
Forward to the second input of LCM 930, 935 and 940
You. FIG. 10 is a block diagram showing the SU 105 according to the present invention.
FIG. SU105 is serial / PCMCIA / Pilot
Interface 1000, multiple TCP / UDP ports
Physical channel mapping module (TUPLCM) 1005,
Mobility management module 1010, security management
Module 1015, logical channel module 102
0, 1025 and 1030, logical channel multiplexing module
Module 1035, modulation / coding / crest factor reduction symbol (ME
CFR) 1040, RF transceiver (RF XCVR) 1050
And 1055, OFDM baseband processing module 104
5 and 1060, multi-antenna coupling module 10
65, detector / receiver 1070, and logical channel
Equipped with a demultiplexing module 1075
You. Serial / PCMCIA / Pilot interface
Interface 1000 supports the RS-232 serial port.
Physical interface, PCMCIA card, 3com
Ilot interface or other similar interface
Interface with the IP packet data stream.
A ream 1003 is generated. TUPLCM 1005 is an IP packet data stream.
And receive it as LCM 1020, 1025
And one of 1030. TUPLCM1005 is LC
M1020, 1025 and 1030 are
Is associated with the UDP port number. Mobility management module 1010 notices
 (Care-Of) Map IP address to real IP address
And make sure mobile IP is supported. security
The security management module 1015 performs authentication, and
That the subscriber trying to access the
confirm. Each LCM 1020, 1025 and 1030
Has a first bidirectional port connected to the TUPLCM 1005
I have. This interface is specific to TCP / UDP ports.
It supports the default data. Flow into the first port
The data stream is ready for transmission from the second port.
Buffet. Logical channel multiplexing module
Serial data received from port 1035 via the third port
Prepares for transmission from the first port to TUPLCM 1005
Placed in the buffer. Each LCM 1020, 1025 and
And 1030 are delay tolerant transmissions from the first port
The data is sent to the subscriber with the correct Automatic Repeat Request (ARQ) response
Buffer until received. Negative acknowledgment
When (NAK) is received, each LCM 1020, 1025
And 1030 transfer the buffered data to the second port.
Resend from Preferably a delay, such as voice
Sensitive data may not be buffered
You. Each LCM 1020, 1025 and 1030 is soft
It can be realized by hardware. The logical channel multiplexing module 1035
Second port of LCM 1020, 1025 and 1030
Receive data streams from these
Multiplexed onto a single data stream 1038. each
The amount of data retrieved from the data stream is
Data type to match the number of bits in the data burst.
Have been. The MECFR 1040 is the data stream 103
8 and the set of bits contained in the data stream.
Forward errors to group errors and control errors
Difference control (FEC) coded bits and transmitted data symbol
Crest Factor Reduction (CFR) view to minimize peak-to-average ratio
The required dynamics of the SU105
The transmission range is reduced, and the transmission efficiency is improved.
Blocks of coded bits are combined into one,
-Transform (IFFT) is applied to the data stream 1038
And performs orthogonal frequency division multiplexing (OFDM) symbol
Is done. After that, MECFR1040 reports the resulting data
Transfer the trim to RF XCVR 1050. Each of the RF XCVRs 1050 and 1055
Digital input, digital output, and antenna port
It is. Is each RF XCVR 1050 and 1055 an input?
These digital data are converted to analog signals and converted to RF carriers.
And amplify the resulting signal and pass it through a filter.
And transmit from the antenna port. Antenna Po
The data received from the filter is filtered and amplified.
Down-converted, digitized and used from the output port.
Be made available. Each OFDM baseband processing module 104
5 and 1060 are RF XCVRs 1050 and 1055
Receiving digitized symbols from one side and fast Fourier transform
Modulated on each individual subcarrier using a permutation (FFT)
Extract symbols. Each OFDM baseband processing module
1045 and 1060 are available at the output port
The possible resulting data stream is
To be received by module 1065. Above
Using the differential OFDM multi-antenna combining technique
H antenna coupling module 1065 is a single coupled data
Generate the trim 1068. [0141] The detector / receiver 1070
Binary digital conversion of symbols included in the
Stream and convert the CFR bit and error control
Data stream 1073 is obtained. Logic
The channel demultiplexing module 1075
The data burst included in the data stream 1073.
The LCMs 1020, 1025 and 1030
To the third port. [0142] Peer-to-peer connection
Support is easier with WIAS 100
You. As shown in FIGS. 9 and 10, the functions of SU and RP
Are very similar. Open peer-to-peer connection
To begin, an SU, for example, SU 105, may
NC on frequency channel selected according to spectrum management
It just generates a C burst. NCC bursts are especially
Called Peer-to-Peer (PTP) NCC with SU IP address
ing. A second subscriber, e.g.
By registering with SU 105 in the same way that
Can be linked to the SU 105. Like this
Recording, even if a WIAS network does not exist,
A direct data, audio or video link between the participants
Can be. If necessary, WIAS 100 can be used for fiber dispersion.
Data (fiber distributed data -FDD) system
Can be realized. This embodiment is
This is particularly useful for applications in licensed PCS bands.
You. According to this embodiment, the traffic operates asymmetrically.
The ability of the system to adaptively sacrifice, but other FDD
Parallel deployment to the system is facilitated. In this case
WIAS downlink is fixed at 16 slots per frame
(The NCC followed by the 15 downlinks
User data slot). Also, the WIAS uplink is
Fixed at 16 slots per frame (RAS and
Subsequent 15 uplink user data slots
G). WIAS 100 uses a wireless link
Subnets can be connected. For this purpose
There are many products that operate in the high U-NII band that can be used
Existing. A typical example is such a wireless bus.
Is a point-to-point link between the RP and the router.
Deployed using directional antennas to tie
I have. Separately, WIAS100 is a multi-hop (Mult
Can be used in i-Hop) mode. For this
Requires an RF transceiver card as shown in Fig. 9 for the RP
And one for giving subscribers access,
The other is for each multi-hop termination. The above description is for a wire for a campus environment.
Focus on providing less internet access
However, the system of the present invention
Loops, battlefield communications, and
Diversified just like point-to-point data delivery
Wireless framework for existing applications
It is also possible to use it. If you are a savvy
As will be appreciated, the invention is not limited by the scope or spirit of the invention.
Improve and change in various ways unless deviating from God
Is possible. In addition, the present invention considers the description in the specification.
However, if the above-described embodiment of the present invention is implemented,
As will be appreciated by those skilled in the field, the present invention
It is also possible to realize in a form. Description of the specification and
The examples are merely illustrative, and the true scope and spirit of the invention
It is indicated by the full scope of the claims and their equivalents. [Brief description of drawings] In the following, it is included in the specification and constitutes a part of it
Embodiments according to the present invention will be described with reference to the accompanying drawings.
In addition, advantages and principles of the present invention will be described. FIG. 1 shows a wireless Internet access according to the present invention.
FIG. 2 is a block diagram showing a system (WIAS). FIG. 2 shows a subscriber unit in the WIAS system of FIG.
Indicates intra-subnet (intra-subnet) mobility
It is a block and a flowchart. FIG. 3 is a flow chart illustrating a handoff process according to the present invention;
FIG. FIG. 4 shows voice traffic in the WIAS system of FIG.
A block showing different connection scenarios that support
It is a lock figure. FIG. 5 shows a voice trough in the simplified WIAS system of FIG.
Support different connection scenarios
It is a block diagram shown. FIG. 6 transmits multiple traffic types according to the present invention;
FIG. 2 is a block diagram showing a frame structure for performing the above. FIG. 7A shows a transmitter and receiver according to the present invention.
FIG. FIG. 7B shows a transmitter and receiver according to the present invention.
FIG. FIG. 8 is a flowchart showing a transmission / reception process according to the present invention.
is there. FIG. 9 shows the wireless communication in the WAIS system of FIG. 1 according to the present invention.
It is a block diagram showing a port. FIG. 10 is a block diagram of the WAIS system of FIG. 1 according to the present invention;
It is a block diagram which shows an entrance unit.

────────────────────────────────────────────────── ─── Continuation of the front page (72) Inventor Burton Melbourne USA 08876 New Jersey Summerville Riviera Drive 5 (56) References JP-A-9-307526 (JP, A) JP-A 9-93216 (JP, A) JP-A-9-181699 (JP, A) JP-A-8-275221 (JP, A) JP-A-8-275220 (JP, A) JP-A-7-336774 (JP, A) International publication 96 / 022639 (WO, A1) (58) Fields investigated (Int. Cl. ⁷ , DB name) H04B ^7/ 24-7/26 H04Q ⁷ /00-7/38

Claims (1)

(57) Claims 1. A system for decoding symbols modulated on a plurality of subcarriers, each symbol corresponding to one or more bits, wherein: Each combination of bits represents a unique phase, and means for receiving from the plurality of antennas a waveform formed from a plurality of antennas and a plurality of superimposed modulated subcarriers, each modulated subcarrier having a different frequency. And means formed by modulating one of the plurality of serial symbols onto a corresponding one of the plurality of subcarriers based on a phase difference between adjacent symbols of each pair; Means for extracting the response of each of said antennas for each of: and means for forming for each sub-carrier a vector having a plurality of elements equal to the number of said antennas, Each element of the vector of the specific subcarrier is obtained by mathematically combining a unit representing an extracted response of one of the plurality of antennas for the specific subcarrier, and a vector corresponding to the adjacent subcarrier of each pair. Means for calculating the phase difference between adjacent subcarriers of a pair, and means for determining a value of each symbol based on the phase difference obtained from the mathematical combination of adjacent vectors of each pair. Features system.