Multiplex communications – Communication over free space – Having a plurality of contiguous regions served by...
Reexamination Certificate
1999-07-07
2004-11-30
Pham, Chi (Department: 2667)
Multiplex communications
Communication over free space
Having a plurality of contiguous regions served by...
C370S343000, C370S345000, C370S436000, C370S437000, C370S468000
Reexamination Certificate
active
06826160
ABSTRACT:
TECHNICAL FIELD
The present claimed invention relates to the field of network communication. Specifically, the present claimed invention relates to an apparatus and a method for dynamically allocating time slots in multiple channels of a bandwidth over which a device may communicate in a network.
BACKGROUND ART
Network communications are ubiquitous in government, businesses and education. Many obstacles have arisen in the effort to better utilize network resources, to provide better quality service to users, and to provide flexible service options to users. One type of network uses multiple frequency channels over which devices may communicate. This type of network is referred to as a Broadband Access Network (BAN). Several of the obstacles for optimal utilization of a BAN will be presented along with the consequential need to solve them.
Prior Art
FIG. 1
a
presents a block diagram of a conventional shared media network system. For exemplary purposes, four devices, e.g. Customer Premises Equipment (CPE),
104
a
,
104
b
,
104
c
, and
104
d
are coupled to a hub
102
via link
106
. Link
106
can provide multiple channels of communication, typically dependent upon the transmitter/receiver capabilities of devices
104
a
,
104
b
,
104
c
, and
104
d
and the hub
102
.
There are two basic methods of communicating in the present network, contention access, and reservation access. With reservation access, a device sends a request to a hub to reserve a specific length of transmission on the network. The hub processes the request, reserves a slot on the network for the requesting device, and sends instructions to the device telling it when to transmit. Subsequently, the device transmits its information according to the instructions. This process avoids collisions, but it also consumes time and resources to make and process the request. Delay sensitive applications such as Internet Telephony and video conferencing require the quality of service delivered by reservation access protocol.
On the other hand, data transmissions such as e-mail and file transfers can be more efficiently delivered by contention access protocol. With contention access, any device in a network can communicate when it has information ready to transmit and when a unused time slot appears. Unfortunately, if more than one device meets the same criteria and tries to transmit at the same time, a collision may arise. If a collision arises, the hub will send a message to the relevant devices indicating that the transmission has failed. Consequently, any device involved in a collision will have to attempt to retransmit the information. In broadband access networks similar to the one shown in Prior Art
FIG. 1
a
, the multiple access protocol is generally a combination of reservation access and contention access for Internet access and Telephony services. A separate downlink channel is available for the hub to communicate to the devices. However, there are much fewer problems with the downlink channel than the uplink channel because a single entity, the hub, controls all transmissions on the downlink.
Prior Art
FIG. 1
b
provides an illustration of multiple uplink channels (channels) that can be used in a hub-spoke network. It also exemplifies a conventional assignment of time slots in multiple channels over which multiple devices will communicate. The illustration presents a contention access scenario where devices communicate and hope to avoid a collision. In the example, three channels, channel AA
154
, channel BB
152
, and channel CC
153
are shown in parallel as they span the spectrum
164
of the network. They represent different frequency channels within the bandwidth, or spectrum
164
, of the network. As shown, several packets fill up lower channel AA
154
. These packets are sequentially identified as a first data packet
156
a
for device A
104
a
, a data packet
158
for device B
104
b
, a second data packet
156
b
for device A
104
a
, and a data packet
160
for device C
104
c
. This scenario illustrates that the efficiency of contention access drops when a large number of time slots in a shared channel are assigned to network devices by the reservation protocol.
Given the scenario in Prior Art
FIG. 1
b
, a first attempt to transmit a data packet
162
a
for device D
104
d
, and a second attempt to transmit a data packet
162
b
for device D
104
d
, occur because device D
104
d
cannot locate an available time slot in which to transmit. Hence, device D
104
d
has an extended wait time for communicating, and a lower quality of service. Consequently, a need arises for a method of allowing a fairer and more consistent access to the network by devices.
One of the conventional methods of providing access to the network when one channel becomes very busy is to force the device, whose transmission is being blocked, to migrate to an empty uplink channel. In the present example, device D
104
d
can migrate up to channel BB
152
after repeatedly being blocked in its originally allocated channel, channel AA
154
. However, this process is inefficient. Channel resources may be overburdened in one channel and underutilized in another channel. Additionally, the waiting period for the blocked device and the consequent migration process consumes valuable time and results in substandard communication performance. Consequently, a need arises for a method to more evenly utilize the multiple channel resources and thereby provide better communication performance to the user.
A user will typically transmit in just that given channel for consecutive data packets. Even if a user is forced to migrate to another channel, the user will transmit just in the new channel to which it migrated. However, sometimes two or more channels are equally busy or loaded, and assigning a device to transmit entirely on one or the other channel will cause congestion. Hence, a need arises for a method and apparatus to allow a new device to transmit on evenly loaded channels without upsetting the even loading.
As illustrated in Prior Art
FIG. 1
b
, a device is conventionally allocated a default channel on which to transmit. This default channel conventionally remains the same for a given device. Even though different devices may be assigned to different channels in the multichannel BAN, they still maintain their default channel each time they access the network. Hence, in each channel, the bandwidth utilization can vary at any given time depending on the number of devices that are active or idle on that channel. Consequently, different channels in a multichannel network can have vastly different loading depending on whether, or when, devices log on to their allocated channel. Furthermore, the loading in each channel can be highly non-uniform if a device with a low bandwidth capability is grouped into a channel having a large bandwidth. The varied loading might mean overcrowding on one channel and unused resources on another channel.
In the present example, device A
104
a
, device B
104
b
, device C
104
c
, and device D
104
d
are all allocated to the same default channel, channel AA
154
. Alternatively, these devices could have been assigned to a different default channel. The point is that, regardless of which channel they are on, once a device is assigned to a channel, the assignment does not change. As a result of this rigid convention, a need arises for a method to periodically allocate default channels in a dynamic manner to better utilize the resources on a multichannel network.
Another feature of the conventional protocol is to assign a device just a single fixed channel in which to transmit information. In conventional networking, more than one service class can be supported on a single network. However, the conventional method does not accomplish this service efficiently. The conventional method will create channel bandwidths that are specific to a service class of user.
Prior Art
FIG. 1
c
provides an illustration of multiple uplink channels (channels)
170
for different service classes that can be used in a hub
Wang Peter S.
Yu Ben Y.
3Com Corporation
Ly Anh-Vu
Pham Chi
LandOfFree
Dynamic bandwidth allocation through multi-channel time slot... does not yet have a rating. At this time, there are no reviews or comments for this patent.
If you have personal experience with Dynamic bandwidth allocation through multi-channel time slot..., we encourage you to share that experience with our LandOfFree.com community. Your opinion is very important and Dynamic bandwidth allocation through multi-channel time slot... will most certainly appreciate the feedback.
Profile ID: LFUS-PAI-O-3335833