Note: Descriptions are shown in the official language in which they were submitted.
CA 0221~07~ 1997-09-09
SYSTEM FOR PROVIDING STREAM BASED
AND PACKET BASED SERVICES
Field of the Invention
The present invention relates to CDMA networks and to integrating
the provision of stream-based services (eg., voice) and packet-based
services.
10 Background to the Invention
In wireless systems, there exists a need to support a wide variety of
multimedia services, such as voice, data, video, facsimile, and interactive
services, simultaneously for many different users. The users, in turn, have
15 differing requirements. For example, voice or other real time calls are
examples of stream based services which require dedicated
communication channels, as lengthy delays in transmission are not
acceptable. However, certain data calls, for example electronic mail or other
text messages, do not require dedicated communication channels as
20 variable or lengthy delays in transmission can be accommodated.
In a conventional IS-95 CDMA (code division multiple access)
system, packet data services are provided through traffic channels, which
were originally designed for voice traffic. A packet data call request and the
25 responses to the request are communicated through the paging and
access channels. Then, a traffic channel is assigned to the packet data call
in a circuit-switched mode. This method causes significant transmission
overhead when the packet data traffic is bursty. A further drawback is that
ATM cells cannot be transmitted directly in this system.
CA 0221~07~ 1997-09-09
Accordingly, a need exists for a novel system structure and method to
more efficiently and more reliably accommodate third generation wireless
systems which encompass various services1 including variable rate packet
data services.
Summary of the Invention
One aspect of the invention provides a novel common packet data
channel (CPDC) system structure and method to efficiently provide variable-
10 rate packet data services in a CDMA system. This system uses CPDCchannels to serve packet data calls for a plurality of packet data users within
a cell or sector. CPDC channels are defined for forward link (base station-
to-terminal) channel and a reverse link (terminal-to-base station)
communications. Preferably, on the forward link (FL), an ATM-type
15 multiplexing scheme is employed on a CPDC channel, while a spread
ALOHA-type random access scheme is used on the reverse link (RL).
Advantageously, the overhead and delays due to call setup and channel
assignment in a conventional circuit-switched system is reduced for
packets in this system. Overhead of idle transmissions for inactive
20 terminals is also reduced. The preferred embodiment allows ATM cells to
be directly transmitted over the air interface.
One advantage of CDMA systems compared to other types of digital
cellular systems is that several terminals can communicate using the same
25 frequency simultaneously. It is well known for CDMA systems that several
virtual channels can be used by having each communication unit code their
transmissions with a distinct spreading code sequence. Furthermore, only
a communication unit with the appropriate code can decode such a coded
message. This provides security against eavesdropping and also provides
30 a multiple access scheme. These aspects of CDMA are well known in the
art. It is also known that several communication units can communicate
CA 0221~07~ 1997-09-09
using the same spreading code as long as they are offset in time. This is
currently utilized in the access channel, which is shared by a plurality of
terminals in a cell/sector. Thus if, for example, three terminals send
requests on the access channel using the same spreading sequence, the
5 first received request can be successfully decoded, even if they overlap,
provided they are offset in time. Of course in conventional systems, the
subsequent requests are discarded. However, we have taken advantage of
this property of a CDMA system to allow a plurality of signals overlapping
(but offset in time) in the reverse link. Thus, a plurality of terminals can send
10 overlapping transmissions of packet data using the same spreading code
sequence to the base station. A base station according to a preferred
embodiment of the invention can decode a plurality of overlapping frame
signals by utilizing a plurality of CPDC receivers which each decode a
separate signal. Thus, for example, a base station can receive multiple
15 packet messages from a plurality of terminals all using the same spreading
code sequence as long as there is some time offset between the reception
of the plurality of signals by the base station.
Another aspect of the invention provides a terminal (and method of
20 using said terminal) for using said CPDC system.
Brief Description of the Drawings
The present invention, together with further objects and advantages
25 thereof will be further understood from the following description with
reference to the drawings in which:
Figure 1 is a schematic drawing illustrating the channel structure in a
CDMA system for providing both stream and packet based services
30 according to a preferred embodiment of the present invention.
CA 0221~07~ 1997-09-09
Figure 2 is a schematic drawing illustrating the packet data control
broadcast channel packet structure according to the embodiment of
Figure 1.
Figure 3 illustrates the forward link common packet data channel
(CPDC). Figure 3a illustrates the logical structure of the packets and Figure
3b illustrates the frame structure of the channel as transmitted by the base
station.
Figure 4 illustrates the reverse link common packet data channel.
Figure 4a illustrates the logical structure of the packets and Figure 4b
illustrates the frame structure of the channel as transmitted by the terminal.
Figure 5 is a flowchart of the steps carried out by a terminal according
to a preferred embodiment of the present invention.
Figure 6 is a flowchart of the steps carried out for an outgoing data
call by a terminal according to the embodiment of Figure S.
Figure 7 is a state flow diagram illustrating the steps carried out by
both the terminal and the base station for an incoming data call according to
a preferred embodiment of the present invention.
Figure 8 is a block diagram of a base station CPDC transceiver
architecture according to a preferred embodiment of the present invention.
Figure 9 is a block diagram illustrating the transceiver architecture of
a terminal according to a preferred embodiment of the present invention.
Figure 10 is a block diagram illustrating the preferred error correction
coding and decoding procedures.
CA 0221~07~ 1997-09-09
Figure 11 is an alternative embodiment of Figure 6.
Detailed Description of the Preferred Embodiments
A preferred embodiment of the present invention will now be
discussed with reference to an IS-95 CDMA system. However, the
principles can be applied to wideband CDMA systems and multi-carrier
(MC) CDMA systems.
We will also describe the preferred embodiment with reference to a
multi-mode terminal which can communicate using stream based or packet
based services or both services simultaneously. Note that terminals can be
voice only, multi-mode, or packet only in this system. Furthermore, the
15 preferred embodiment of the present invention will be described with
respect to mobile station terminals. However, it should be noted that the
terminals may take other forms, including for example, fixed wireless
access terminals and portable computers using a CDMA transceiver.
Figure 1 shows the channel structure in an omnidirectional cell or a
cell sector. The straight lines with downward and upward arrows are CDMA
channels on the forward and reverse links, respectively. By "channel", we
mean a logical channel representing data on a CDMA carrier which can be
decoded using a particular spreading code sequence. For example, there
is a separate spreading code sequence for each of the pilot,
synchronization, paging and access "channel" functions. The pilot, sync,
paging and access channels function in the same manner as in IS-95. As
can be seen in Figure 1, there can be multiple paging, access, stream traffic
and forward and reverse link CPDC channels. There could also be multiple
PDCB channels. Figure 1 also illustrates a base station serving a plurality
of terminals using the stream traffic channels, as well as a plurality of
CA 0221~07~ 1997-09-09
terminals using the packet data services (that is to say the CPDC channels)
in either an omnidirectional cell or a cell sector. The stream traffic channels
provide the circuit-switched services including voice and data services,
which involve transmission of a large amount of information over a relatively
5 long time duration, such as video, facsimile, and long data file transfer. Thestream traffic channels are organized in pairs as each "call" on a stream
traffic channel comprises a forward link from the base station to the
terminal, and a reverse link from the terminal to the base station. Each link
is assigned a CDMA spreading code in the conventional manner.
The packet data control broadcast (PDCB) channel and the forward
and reverse link common packet data channels are used to provide packet
data services. The packet data control broadcast channel is assigned a
distinct code sequence, which is known to all packet data users within a
15 sector. Each forward link and reverse link CPDC channels is also assigned
a distinct code sequence which is used by a plurality of terminals. To
achieve duplex communication, reverse link and forward link channels will
often be organized in pairs. It should be appreciated that some data
applications will not require full duplex communications, for example an e-
20 mail message. Even for such simple services, transmission of responses,such as acknowledgements and other control information are typically
required. These responses may be transmitted on a "paired" channel,
although this is not necessarily the case. For example, responses for one-
way reverse link data packets from a terminal to the base station can be
25 sent on the PDCB rather than a "paired" FL CPDC.
When a terminal registers, the base station sends registration
information to the terminal which specifies which spreading code
sequences to use for both the PDCB channel, and the reverse link CPDC
30 channels. This information can include parameters to generate the
spreading code sequences as a notification as to which previously defined
CA 0221~07~ 1997-09-09
sequence the terminal should use. The terminal stores this information in
its internal memory, and uses the specified code sequences to monitor for
packet page messages from the PDCB channel and for transmitting
packets on the reverse link CPDC. Typically, a packet page message on the
PDCB channel will provide information which specifies which spreading
code sequences to use to receive incoming packets from a forward link
CPDC. However, such information could also be provided on registration.
The control and signalling information related to packet data services
10 is conveyed through the packet data control broadcast channels to all packet
data terminals within a cell or cell sector. Figure 2 is one possible structure
of the packet data control broadcast channel. Each packet comprises a
header section, an information section and a CRC code. The header
section comprises addressing information directed to a particular terminal
15 (or a broadcast message to a group of terminals or all terminals), control
information flag and a CRC code. The two CRC codes are used by the
terminal for error detection purposes. The control information flag is used to
indicate whether the payload is data, control information (and, if so, of which
type) or null. These data packets typically undergo forward error correction
20 coding, interleaving, modulation and spreading to form a frame structure
(not shown) which is transmitted. The control information can include:
packet page message;
radio link protocol (RLP) ARQ acknowledgement;
CPDC channel assignment message on forward and reverse links;
and packet data rate;
and preferably can include:
power control message;
load balancing message;
congestion control message; and
flow control message.
CA 0221~07~ 1997-09-09
Examples of data in the PDCB include short messages for a packet
data capable terminal (note that short message service (SMS) messages
can still be sent in the conventional manner using the paging channel, but
including such messages in the PDCB can allow packet-only-terminals to
5 receive the equivalent to SMS).
A base station may have one or more forward CPDC channels per
sector or omnidirectional cell. Each forward CPDC channel is assigned a
distinct code sequence which is shared by a plurality of terminals.
10 Preferably, the base station determines and adjusts the transmission rate
according to the packet data traffic, the intetference levels of the system, theavailable system capacity and the priority of the ongoing packet data call(s).
The information on the transmission rate is broadcast on the packet data
control broadcast channel. Data packets destined to all or some of the
15 packet data users within the sector are transmitted through a forward CPDC
channel. The data packets to be transmitted are buffered at the base station.
Typically, the base station, as it serves a plurality of terminals, will be
transmitting a continuous stream of packets. Preferably, the base station
pipes the packets through a forward CPDC channel by using an
20 asynchronous transmission scheme such as ATM. The base station
arranges the transmission order of the data packets based on certain
priority criteria (for example, quality of service) or on a first-come-first-served
basls.
Figure 3 illustrates a forward common packet data channel structure.
Figure 3 comprises Figure 3a, which illustrates the logical packet structure1
and Figure 3b which illustrates the frame structure of the channel. Referring
to Figure 3a, each packet in the forward CPDC channel has a fixed length in
bytes and comprises a header section, a payload section and a CRC code.
The payload can carry user data, control information or null data. Preferably,
the payload is sufficient to contain the payload of one or more ATM cells.
CA 0221~07~ 1997-09-09
The header section comprises address information for which terminal the
packet is directed, control information, packet sequence number information
and a CRC code to be used for error detection/correction. Preferably, the
data section is encrypted for security purposes. Figure 3b will be discussed
5 below.
Note that a packet in the PDCB channel in Figure 2 is shown to have
a fixed length in time, whereas the packet in Figure 3 is stated to have a
fixed length in bytes. This indicates that the PDCBis broadcast at a
10 constant rate, whereas the rate of broadcast of the CPDC packets is
variable.
A base station may have one or more reverse CPDC channels for a
sector or omnidirectional cell. A reverse link channel is shown in Figure 4.
15 Figure 4 comprises Figure 4a, which illustrates the logical packet structure,and Figure 4b which illustrates the frame structure of the channel. Referring
to Figure 4a, each reverse CPDC channel is assigned a distinct code
sequence which is shared by a plurality of terminals. Each terminal may
transmit at varying data rates, which are determined by the base station and
20 transmitted to a terminal by means of a PDCB or FL CPDC packet data rate
or flow control message. Note that the terminal preferably starts
transmission of its packet burst at a default rate in order to conserve
network resources and then the terminal can adjust the transmission rate if
it receives a flow control message from the base station. Note that if there
25 are multiple CPDC channels, they may have different transmission rates. A
packet data capable terminal originates a packet data call or burst by
starting transmission whenever it has messages to be delivered to some
destination (unless the terminal has received a congestion message via the
packet data control broadcast channel). Thus, the CPDC reverse link
30 employs an ALOHA-type random multiple access scheme. For the purpose
of synchronization acquisition, a preamble is transmitted at the beginning of
CA 022l~07~ l997-09-09
a packet data burst. The preamble also includes control information,
including the terminal ID and acts as a request by the terminal to the base
station to transmit packet data, and/or the control information can include
acknowledgement data, in which case the preamble acts as an
5 acknowledgement message.
The forward and reverse link CPDC's effectively give to the terminal a
virtual circuit because all packets directed to the terminal can be received by
the terminal over the forward link CPDC and whenever the terminal has a
10 packet to send it can transmit the packet or burst of packets on the reverse
link CPDC (subject to a congestion message broadcast from the base
station). Thus, as far as the terminal is concerned, a virtual circuit for
providing packet data communication is established between the base
station and the terminal, even though a plurality of other terminals can be
15 using the same forward and reverse link CPDC channels at the same time.
Furthermore, the terminal can utilize more than one virtual circuit if needed
for multiple applications.
Each packet in the reverse CPDC channel has a fixed length in bytes
20 and comprises a header section, a payload section and a CRC code. The
payload can carry user data, control information (for example, ARQ/ACK and
radio link protocol (RLP) control message) or null data. Preferably, the
payload is sufficient to contain the payload of an ATM cell. The header
section comprises source and destination address information, control
25 information, packet sequence number information and a CRC code to be
used for error detection/correction. Preferably, the data section is encrypted
for security purposes. Figure 4b will be discussed below.
Figure 5 is a flow chart illustrating the steps carried out by the
30 controller of the terminal. When the terminal first powers up 1000, the
terminal enters into an initialization state 1050. During the initialization
CA 0221~07~ 1997-09-09
11
state, the terminal boots and then scans the pilot and sync channels in a
conventional manner. Once the terminal has acquired the pilot and system
timing, the terminal performs registration 1100, where conventional
registration tasks are performed, and the terminal also receives and stores
5 spreading code sequences 1150 in order to access and use the various
system channels. In particular, the terminal receives registration
information which specifies the spreading code sequences for the packet
data control broadcast channel and the common packet data channels as
discussed above. As noted previously, the CPDC channels comprise both
10 forward link and reverse link CPDC channels. As previously stated, a
particular cell or sector may have a plurality of CPDC channels. In such
circumstances, upon registration, the terminal receives from the base
station an indication via the PDCB as to which of the RL CPDC channels to
use for page response messages referred to in Figure 7, as well as which
15 RL CPDC to use to originate packet calls. The terminal then enters the idle
state 1170. During the idle state, the terminal monitors the network for an
indication that an idle hand off is appropriate, monitors its own internal
operations to determine whether an outgoing call is to be made, and also
monitors both the paging channel and the PDCB channel for an incoming
20 stream call page or an incoming packet page message. Hand off 1180 is
handled in the conventional manner. An outgoing "call" can be handled in
one of two ways. It can be handled by making a stream call or it could be
handled by sending data packets where appropriate. According to the
preferred embodiment, the mobile station first makes a decision as to
25 whether it is a voice call in which case it is handled by a stream based callin a conventional manner 1400. If it is a data call, the call could be handled
either in a stream based service or as a packet, depending on the nature of
the data to be sent (and possibly also by the capacity on the stream services
channels and/or the packet data channels).
CA 022l~07~ l997-09-09
12
When the terminal needs to set up a data call 1500, the terminal first
determines whether the data is best suited for packet delivery or whether it
requires a dedicated streaming service (circuit switched call) 1510. Note, in
this example, we are assuming the terminal can communicate by voice or
5 by streamed or packet data, for example, a fixed wireless access subscriber
unit capable of providing telecommunication services to both telephones
and modems for transmitting data from a personal computer, or a wireless
web browser phone capable of receiving and using data as well as
providing voice telephony. Table 1 lists a series of possible services
10 available to the terminal, and the preferred type of communication service
required for each.
Table 1
Teleservice Service Type
Speech circuit switched
Asynchronous Data circuit switched
Rate Adapted ISDN circuit switched
Group 3 Fax circuit switched
Video Telephony circuit switched or packet data
Electronic Mail packet data
Database Access packet data
World Wide Web Access packet data
File Transfer packet data
Broadcast Data packet data
Program Sound packet data
Electronic Newspaper packet data
15 In such a situation, the enquiry at 1510 can be answered by an indication by
the user/computer application as to the type of service required.
(Alternatively, a sophisticated terminal can analyse the type and length of
data in order to determine the type of service required. This will require
CA 022l~07~ l997-09-09
13
some buffering and/or possibly redirection of the call after some interval.)
Assuming the first option, let us take as an example a fixed wireless access
subscriber unit which acts as a telephone and a modem for a personal
computer. The subscriber unit can determine whether the handset for the
5 telephone has gone off hook, and in which case, use the access channel to
set up a circuit switch stream service for voice communications. For
situations where a user wants to send data, the computer can be equipped
with software for sending an indication to the terminal as to the type of
request being made. In this example, step 1510 is answered by either an
10 indication from the user or software resident in the computer for directing
data files according to Table 1. For example, let us assume that the user of
the terminal has requested to send a document by fax. In which case,
according to Table 1, a fax is best handled by a streaming service. The
terminal will receive such an indication from the computer and then select
15 the streaming service call processing step 1520. Let us now assume that
the user of the computer has made a request to send a text message by
electronic mail. Such a text message is best treated as packet data and the
computer will therefor instruct the terminal that a packet data call is requiredand establish virtual circuit 1515. Subsequent e-mail messages are
20 directed to this virtual circuit for transmission of data packets. Of course, the
terminal and computer can be integrated into a single unit. The steps for an
outgoing data call, starting at step 1515, will be discussed below with
reference to Figure 6.
Continuing with the idle state according to the preferred embodiment
in Figure 5, with respect to incoming calls, the terminal is equipped with two
code sequence generators so that it can monitor the paging channel 1600
and the PDCB channel 1650 in parallel. If an incoming streaming services
page is detected at 1610, the terminal proceeds to conventional stream
service call set-up 1620; otherwise, the terminal continues to monitor for
incoming pages. Meanwhile, the terminal is monitoring the PDCB for
CA 022l~07~ l997-09-09
14
incoming packet page messages at 1650. If such a page is received 1660,
the terminal processes the incoming packet page according to Figure 7. If
no such packet page message is received, the terminal continues in the
idle state to monitor incoming pages. Note that the terminal can monitor for
5 pages in an alternative manner. For example, if only one code sequence
generator is used, the terminal can monitor the paging channel and the
PDCB channel in a serial alternating manner rather than in a parallel
manner. As a further alternative, a terminal can only monitor the PDCB if the
base station is configured to send both stream and packet page messages
10 via the PDCB. Note that for battery powered terminals, in order to conserve
battery power, the terminal may enter a sleep mode during the idle state. In
which case, the terminal will "wake up" periodically, for example once per
second, to carry out the idle state processes.
Referring to Figure 6 when packets are delivered to the terminal for
transmission 1515, the terminal monitors the PDCB to determine whether a
congestion message has been transmitted by the base station 1530. A
congestion message may take the form of a specific message transmitted
on the PDCB. Alternatively, a set of congestion control bits can be sent in
20 routine messages. If a congestion message has been received, the
terminal waits a random time delay and then re-checks the PDCB for a
congestion message 1540. Assuming no congestion message has been
received, the terminal proceeds to transmit. Note, as an alternative, the
terminal can constantly monitor the PDCB and set a flag whenever such a
25 congestion message is received. The terminal will also flag whenever the
PDCB is not congested. In which case, step 1530 simply involves
consulting the flag to make the determination. In any event, assuming no
congestion, the terminal transmits the preamble on the reverse link CPDC
1560. The preamble comprises sync information to the base station as well
30 as a control information which includes the terminal ID and identifies to the base station the terminal making a request to send data. The terminal
CA 0221~07~ 1997-09-09
starts an acknowledgement timer 1550, and then at step 1560, the terminal
transmits the preamble on the reverse link CPDC as described with
reference to Figure 4. The terminal then monitors the PDCD at 1570 for an
acknowledgement from the base station of the preamble transmitted at
1560. While no acknowledgement exists, the timer continues 1575 and
unless the timer has expired, the terminal continues to retransmit the
preamble 1560. If the timer has expired, the terminal then waits a random
period of time and attempts to transmit again 1580. As soon as the terminal
receives the acknowledgement from the base station, the terminal transmits
10 the burst of packet data according to the structure of Figure 4. While
transmitting the packet 1590, the terminal performs radio link protocol (RLP)
tasks, for example, it monitors the PDCB for acknowledgements by the base
station and retransmits packets if needed. As an alternative, the
acknowledgement can be sent by the BTS on the forward link CPDC as a
15 control packet directed to the terminal or as subchannel information as
discussed below. The burst is variable in length as the terminal continues
to transmit packets of data until no additional packets (or retransmitted
packets) are waiting to be sent. The terminal can also transmit subchannel
information as discussed below with reference to Figure 4b.
Figure 7 illustrates the steps taken by both a terminal and its serving
base station, and the messages sent between them when a packet "call"
terminates with the terminal. In such a situation, the base station will
receive a terminal termination request from a network switching element, for
25 example, a mobile switching centre. The base station detects a terminal
termination call request and sends a packet page message directed to the
terminal via the PDCB channel. This page message will typically
information which specifies which spreading code sequence (i.e., which
reverse link CPDC channel) that the terminal is to use to send
30 acknowledgement messages unless it is a previously arranged default
channel. The terminal will then send a page response message on the
CA 022l~07~ l997-09-09
16
reverse link CPDC, acknowledging the packet page message, according to
the radio link protocol used. Once the base station receives the page
response message from the terminal, a virtual circuit will be assigned
between the terminal and the base station, if one has not been assigned
5 already for a particular application, by sending a CPDC assignment
message to the terminal assigning a forward link CPDC to be used for this
virtual circuit. The terminal will receive the assignment message and turn to
the assigned forward link CPDC in order to monitor for incoming
messages. The base station then transmits messages according to the
10 frame structure of Figure 3b which includes data packets and subchannel
information directed to the terminal on the assigned forward link CPDC.
The terminal receives the transmitted frame structure on the forward link
CPDC and receives the data packets and subchannel information directed
to it. At this point, it should be noted that the terminal receives, despreads
15 and demodulates the entire frame structure in order to process the data
packets and subchannel information directed to it. Data packets and
subchannel information directed to other terminals on the FL CPDC are
discarded. In so doing, the terminal performs block deinterleaving and error
control decoding. Error detection is performed in the terminal which sends
20 acknowledgement messages according to the ARQ protocol on the
designated reverse link channel. As stated, these acknowledgement
messages can be in the form of packets transmitted on the reverse link
CPDC channel as described herein or preferably the preamble can contain
the acknowledgement message. In any event, the base station receives the
25 appropriate acknowledgement messages and acts accordingly, for example
retransmitting packets if needed.
Figure 8 is a block diagram showing the base station transceiver
architecture for handling CPDC traffic. It should be recognized that many
30 other components for handling the full functionality of the base station
transceiver are not shown. Communications between terminals and the
CA 022l~07~ l997-09-09
17
base stations occur via the antenna 100 and radio unit 110. Note that a
plurality of radio units may be utilized. The RF unit 110 handles
conventional radio functions for both receiving and transmitting, for example,
antenna duplexing, power amplifying, down and up converting, etc. For
5 CPDC purposes, the base station also comprises a control unit 135, and a
plurality of CPDC receivers (preferably there will be one CPDC receiver for
each terminal transmitting on the CPDC channel). Two such CPDC
receivers 200 and 205 are shown. The base station also includes a CPDC
transmitter 500 and a channel combining and spreading unit 700.
When a packet burst from a reverse link CPDC channel is received by
the base station, it is down converted by the RF unit 110 wherein the down
converted signal is sent to each of the CPDC receivers and the control 135.
The control unit 135 comprises a preamble detection and synchronization
15 acquisition block (for example a DSP) 130 as well as a controller 140 which
includes, for example, a microprocessor and associated memory. The
preamble detection and synchronization acquisition unit 130 continuously
searches for reverse link CPDC for preambles sent from the terminals. It
should be noted that the preambles will all follow a common format and
20 hence can each be detected by the single unit 130 using conventional
methods used for detecting preambles in reverse link access channel
messages. As stated earlier with reference to Figure 4, the preamble can
act as a request by a terminal which has data packets to transmit, or it can
include an acknowledgement message from the terminal (or it can do both).
25 If the preamble only includes acknowledgement information, the controller
140 can process the acknowledgement directly without processing by a
CPDC receiver.
The unit 130 captures the preamble and performs synchronization
30 acquisition and then sends the timing information and control information to
an available CPDC receiver, for example, receiver 200. The controller 140
CA 022l~07~ l997-09-09
18
allocates each separate incoming packet burst from a particular terminal to
an available CPDC receiver. To do this it maintains records of which CPDC
receivers are available at any instant in time. If all CPDC receivers are busy,
the control unit 140 sends a message to the PDCB unit (not shown) to
5 transmit a congestion message on the packet data control broadcast
channel. Assuming CPDC receiver 200 is available, the control unit 135
sends the timing information to the despreading and demodulation unit 220
of the allocated CPDC receiver. Note, as described no physical switching is
necessary. Each CPDC receiver is connected to the RF unit and is capable
10 of despreading the packets. The unit 135 allocates a receiver by sending
the timing information necessary to decode the particular packet burst to the
allocated CPDC receiver. Once allocated, the CPDC receiver continues to
process frames received from the transmitting terminal until the burst is
completed. Note that although not necessary, as described some switching
15 of the CPDC receivers may be desired for other purposes, for example
concentration of CPDC receivers.
Thus, this particular base station architecture takes advantage of the
fact that as long as there is some time offset between packet bursts on the
20 RL CPDC, each CPDC receiver can decode a particular terminal's packet
burst. In despreading and demodulation block 220, the received signal is
despread using the reverse link CPDC spreading code sequence and, then,
the despread signal is demodulated in a conventional manner. The output
signal, a digital bit stream in the frame structure as shown in Figure 4b, is
25 fed to the demux block 240 of CPDC receive frame/packet processor 230. At
this point, the reader should note that each receive frame/packet processor
230 is shown to carry out several block functions. The reader should note
that although several block functions are shown, these can be grouped or
performed in different ways either by discrete circuits, a microprocessor with
30 associated software and memory, or other integrated circuits, for example
CA 022l~07~ l997-09-09
19
DSPs. In any event, the blocks can be considered to include instructions to
a controller for carrying out the blocked function.
In the demux block 240, the multiplexed subchannel information; for
5 example, power control information (power control flag and power control
bits) are extracted from the frame and processed. If power control
information is present (eg. the PC flag is on), the PC information is fed to thepower level adjustment block 260. The power level adjustment is then sent
to the base station power control processor which determines the forward
10 link transmit power based on a variety of inputs, including the inputs from
the terminals on the shared CPDC channel and based on the total power
requirements of the cell/section.
Block interleaving and deinterleaving is performed in order to reduce
15 the impact of fading on the channel. Thus, the demultiplexed signal is sent
to block deinterleaving 250. The deinterleaved signal is input to the error
control decoding block 300 where certain transmission errors are corrected.
In order to provide an adequate quality of service, error detection may be
employed to detect non-corrected errors.
Then, the header is abstracted from the decoded signal in demux
block 270 and sent to the header detection block 280 for processing. If the
packet is for control use (for example, a NAK message requiring
retransmission, or a disconnect message), the base station will start
2~ appropriate processes. If it is a data packet without error, the data is sent to
the data decryption block 320. The data decryption block 320 decrypts the
packet and then the data is sent towards its destination, for example by
means of an ATM back haul network from the BTS (or BSC/MSC or other
network node coupled to the BTS).
CA 0221~07~ 1997-09-09
Note that an ARQ protocol is used to improve the integrity of the data.
Various types of acknowledgement messages for example, positive
acknowledgements, or negative acknowledgements (plus some control
signals to indicate end of burst) etc. can be utilized. For example, if the
5 packet is corrupted, a negative acknowledgement will be put into the data
buffer of the transmitter by means of the ARQ/ACK link 295. Alternatively,
such a negative acknowledgement (i.e., a retransmission request) can be
sent to the terminal via the PDCB, as stated above. Preferably, a positive
acknowledgement of the preamble is sent to the transmitting terminal,
10 either via the PDCB or a FL CPDC.
The base station CPDC transceiver architecture also includes CPDC
transmit frame/packet processor 530. Once again, certain block functions
are illustrated which can be executed by discrete circuits or integrated
15 circuits, for example, DSPs or, as shown, a microprocessor with software
stored in an associated memory (not shown). In operation, data input, for
example, ATM cells from an ATM backhaul network, is entered into the cell
buffering and ordering block 520. The payload for each packet is then
encrypted using a suitable data encryption routine via data encryption block
20 540 which also appends a CRC code calculated over the encrypted data,
and then is assembled into packets with an appropriate header by mux 550.
Meanwhile, the header is generated by the header processing unit 525
which carries out the following tasks:
1. The address/ terminal ID field is entered with a virtual circuit
identifier allocated to the destination terminal;
2. The control field flag is specified to indicate whether the
payload is for control use (and if so, can specify the type of
control information) or contains data (or possibly is a null
packet);
CA 0221~07~ 1997-09-09
3. Writes the packet sequence number into the packet number
field; and
4. Computes the CRC check over the header.
The packet is then error control coded by error control coding block
600 and is then fed to the block interleaving unit 570 and interleaved over
one or more frames. Power control or other subchannel information
10 (labelled as PC flag and PC bits in Figure 3b) are then multiplexed within
the frame by means of mux 580. The frame structure for the FL CPDC is
illustrated in Figure 3b. The duration of a frame is typically 10 or 20 ms.
Each PC flag indicates whether power control is on for an associated
terminal over this frame. If PC is on, then the PC bits associated with the
15 PC flag indicate whether the terminal transmit power is to be increased or
decreased. Typically, also, the PC is updated at intervals of 1.25 ms. As
shown in Figure 3b, the FL may support a number of logical PC channels (n
in the figure). Preferably, there is one PC channel per terminal active on the
RL. A specific PC channel is assigned to a specific terminal by a message
20 over the PDCB channel on the FL and an acknowledgement on the CPDC
on the reverse link. If PC is not on, the PC bits, in conjunction with the PC
flag, can be used for other purposes, for example, providing signalling
information. Thus, the frame structure of Figure 3b more generally
illustrates a frame structure wherein the frame includes subchannel
25 information. In this figure, there are n subchannels, wherein each
subchannel comprises a flag and associated bits of subchannel
information associated with each flag, and located at regular intervals within
the frame structure. The flag determines the nature of the associated bits.
Thus, for example, referring to Figure 3b, PC flag 1 could indicate to terminal
30 1 that its associated subchannel bits (i.e., the bits labelled as PC 1 which
are associated with PC flag 1 ) represent power control bits whereas PC flag
CA 0221~07~ 1997-09-09
22
2 could indicate that the subchannel information in associated subchannel
bits (labelled as PC 2) contain other information, for example, flow control
information. Thus, the frame processor 530 produces a frame structure
which includes subchannel information and individual packets for a plurality
5 of terminals.
Continuing with Figure 8, the packets are then subject to Walsh
spreading in a conventional manner via multiplier 590 and then fed to
amplifier 595 the gain of which is controlled by the output from the power
10 control processor (not shown). The signal is then sent to the channel
combining and spreading unit 700 where it is spread according to the
forward link CPDC spreading code and then sent to the radio block 110 for
transmission via antenna 100.
It should be noted that although specific blocks are shown for both
the receiver and transmitter, certain blocks can be combined and shared by
both. For example, each receiver and transmitter are shown to include a
separate header processing or header detection block, whereas they could
share a common header processor which produces the headers for
outgoing packets and decodes the headers for incoming packets. More
generally, although receive frame processor 230 and transmit frame
processor 530 are shown as separate blocks, a single frame processor
which is used to perform both functions can be used.
Figure 9 is a block diagram showing the terminal transceiver
architecture for handling CPDC traffic. It should be recognized that many
other components for handling full functionality of the terminal are not
shown. Furthermore, as with the base station, although several block
functions are shown to be grouped in a specific way, these can be grouped
or performed in different ways, either by discrete circuits, microprocessors
with associated software and memory or other integrated circuits, for
CA 0221~07~ 1997-09-09
23
example DSP. Furthermore, the blocks can be considered to include
instructions to a controller for carrying out the block function.
Many of the blocks and their arrangement in the terminal transceiver
5 architecture follow a similar form as that in the base station architecture.
Therefore we will only describe the significant differences. It can be seen
that in a terminal there will typically be only one CPDC receive frame
processor 830 and one transmit frame processor 850 as one CPDC
transceiver can typically multiplex several data applications at once, by
10 maintaining multiple logical links. Furthermore the control unit for the
terminal (not shown) does not need to perform the preamble detection and
synchronization for a series of incoming packets. The terminal transceiver
will still have a central controller and associated memory (not shown). Of
course depending on the nature of the terminal, more than one CPDC
15 receive and transmit frame processor can be included to increase capacity,
for example, if the transceiver forms part of a wireless interface unit, for
example a fixed wireless access subscriber unit, capable of providing
telecommunications services to more than one terminal unit.
In the preferred embodiment shown, differences between the
terminal transceiver architecture and the base station transceiver
architecture are as follows, bearing in mind the general differences between
base and terminal.
First, the power level adjustment signal 840 produced by the power
level adjustment block is supplied directly to both channel gain amplifier A
865 and channel gain amplifier B 875 for controlling the power of both the
main signal to be transmitted and the pilot signal described below. Power
control information directed to a particular terminal can be directly used by
the terminal to adjust its transmit power based on instructions to it from the
base station. This differs from power control information sent by the
CA 0221~07~ 1997-09-09
24
terminal and received by the base station, as the base station must
consider the PC adjustment requests of all terminals as it regulates the
power control of the entire cell.
Each terminal is shown to include a header detection and processing
block. As noted with reference to the base station architecture, these can be
grouped in a common header processor shared by both the receive
processor and the transmit processor, which performs header detection
and decoding during the receive stage, and header production during the
transmit stage. These blocks perform essentially the same function as
described above with respect to the base station, although the reader
should note that there may be slight differences in the headers due to the
nature of the differences in the data structure between the forward link and
reverse link headers.
Furthermore, the control unit (once again not shown) includes a
memory for storing the spreading code sequences received by the base
station upon registration, which are utilized by the despreading or spreading
blocks to receive and send packets respectively.
Turning now to the transmitter 900 it can be seen that the transmitter
includes a pilot channel 860 with its associated channel gain amplifier 865,
a second channel gain amplifier 875 whose input is selected by switch 872
to be the output from the transmit frame processor 850 or the preamble 870.
A pilot channel 860is provided for coherent reverse link reception at
the base station. The data of the pilot channel is all "+", represented by a
high level signal. The channel gain A is controlled by the signal from the
power level adjustment block in order to obtain the desired transmit power
level. Then, the power controlled pilot channel signal is sent to the 1-
CA 0221~07~ 1997-09-09
channel of the complex spreading block, where the conventional spreading
operation is performed.
The preamble data 870 is produced by a preamble generator (not
5 shown). At the beginning of each data burst, the switch 872 is controlled to
select the preamble for transmission. The preamble data is power
controlled by the channel gain B. The power controlled preamble signal is
sent to the Q-channel of the complex spreading block, where the
conventional spreading operation is performed. When the receiver receives
10 an acknowledgement from the base station, the switch is reset to select the
data transmission branch, wherein transmission of the preamble is
stopped, and transmission of the packets begins. The complex spreading
block 890 performs channel combining and spreading according to the
selected reverse link spreading code sequence. It should be noted that
15 different complex spreading arrangements are possible for the pilot
channel, the preamble and the packet data.
Note that the data input to the transmit frame processor 850 and the
data output from the receive processor is shown to consist of ATM cells, but
20 other formats for the data can be used.
Both the base station and the terminal have shown an error control
decoding and error control coding block. Conventional error control coding
and decoding can be used, however the preferred embodiment of the
25 invention utilizes a specific type of error control coding and decoding as
shown in Figure 10. Note that as an alternative to multiplexing each frame
with subchannel information, subchannel information can be punctured. In
this case, the error control decoding block of the base station or terminal willneed to know the status of the PC flag, for example by a signal (not shown)
30 from the demux block 240. Similarly, on the transmit side, the frame will be
CA 022l~07~ l997-09-09
26
punctured with subchannel information by mux 580 of the base station or
terminal.
For the error control coding, serially concatenated codes with Reed-
5 Solomon (R-S) code as outer code and conventional code as inner code
can be used. The R-S code can be used for both error correction and error
detection for ARQ, as necessary. The R-S code is chosen for the following
reasons. First, the R-S code has strong capability to correct burst errors,
which is desired for an outer code. Second, it is easy to find a R-S code with
10 code symbol of 8 bits, or octet, which an ATM cell consists of, so that an ATM
cell can be directly coded with the R-S code. Figure 10 shows a possible
numerology for the concatenated codes. At the input of the Error Control
Coding block, 52 octets of data, including the header, are fed into the (78,52)
R-S encoder. Code tail bits, for example 8 bits, can be added to the R-S
5 coded data packet in order to be able to decode an inner convolutional code
using only data from this block. Then, the data packet undergoes a
convolutional coding by using a convolutional code of, for example, rate 1/2
with the constraint length of K=9. The convolutional encoded data packet is
then sent to the Block Interleaving block. As can be seen in Figure 10, the
20 reverse steps are carried out during the decoding operation.
For example, we have shown RL and FL CPDC structures wherein
the Power Control (PC) flag is multiplexed within the frame structure. This is
particularly desirable in situations where very quick power control
25 adjustments are needed, as the power control information can be quickly
extracted and used without waiting for error control processing and header
detection. This is assumed in the embodiments shown in Figures 9 and
10, wherein, for example, the Demux unit 240 extracts the power control flag
in order to determine whether power control bits are included (either
30 multiplexed or punctured) and sent for power level adjustment 260, without
having to wait for completion of the header detection block 280. Similarly, on
CA 0221~07~ 1997-09-09
the transmit side, Mux unit 580 multiplexes power control information (which
includes the PC flag and PC bits) into the data as appropriate. However, as
an alternative for appropriate applications, subchannel information, for
example, the PC flag and PC (or other) bits can be included as part of the
5 packet structure.
Figure 11 illustrates an alternative method of transmission of a
packet burst by a terminal. Figure 11 is similar to steps shown in Figure 6,
except that in Figure 11 a terminal does not wait for an acknowledgement
10 from the base station of its preamble before sending packets of data.
Rather, as shown in Figure 11, the terminal transmits the preamble on the
reverse link CPDC and then immediately starts sending packets of data.
These packets of data are organized into frames, which preferably include
subchannel information as described above. While sending the packets of
15 data, the terminal monitors the forward link for both packet errors and a
message acknowledging the preamble. Note that the terminal may be
monitoring both a forward link CPDC and the PDCB at this step. Once the
preamble is acknowledged, the terminal continues to send packets of data
and monitors the forward link for packet errors. The terminal retransmits
20 any failed packet. If the preamble was not acknowledged within the
specified period of time, the terminal assumes that all of the packets of data
which it had transmitted were not received, waits a random period of time
and then retransmits the entire burst.
The preferred embodiments as described above involved the
transmission of frame structures which include subchannel information.
This subchannel information can be used as power control adjustment
information to be used in controlling the transmit power of the receiving unit
(basestation or terminal). Controlling the transmit power of the individual
terminals is relatively straight forward. However for clarity, we will now
elaborate some general principles of basestation power control. There are
two general ways of designing a system as discussed.
CA 0221~07~ 1997-09-09
One way is to simply not use dynamic power control at all. The base
station will transmit at a fixed power level which is preset so that every
terminal in the cell will receive the transmissions, regardless of where they
5 may be located or what local conditions will occur ( within some limits). In
this case, the subchannel information will always be used for some other
purpose, or can be eliminated altogether.
The second way is to dynamically control the power level so as to
10 minimize the output power while providing sufficient power as needed to all
users. This can be done in several ways. For example, the power control
information can be used to provide sufficient power to be received by all
terminals currently using the data channels. Preferably power control is
used on a frame-by-frame basis, wherein each frame is transmitted with
15 sufficient power to reach every terminal for which the frame contains data.
As a consequence of this option, some terminals served by a basestation
may not receive a frame for which there is no data directed to them ( with
sufficient power to properly decode the frame). The basestation should
therefore ensure sufficient frames are received by all terminals using the
20 data channels so as to maintain the communication links.
Numerous other modifications, variations and adaptations may be
made to the particular embodiments of the invention described above
without departing from the scope of the invention, which is defined in the
25 claims.
