Telecommunications – Radiotelephone system – Zoned or cellular telephone system
Reexamination Certificate
1999-10-06
2003-06-03
Appiah, Charles N. (Department: 2682)
Telecommunications
Radiotelephone system
Zoned or cellular telephone system
C455S561000
Reexamination Certificate
active
06574477
ABSTRACT:
CROSS-REFERENCE TO RELATED APPLICATIONS
Not Applicable
STATEMENT REGARDING FEDERALLY SPONSORED RESEARCH OR DEVELOPMENT
Not Applicable
BACKGROUND OF THE INVENTION
1. Field of the Invention
This invention relates to wireless communication networks, including networks servicing cellular telephones, personal communication systems, and the like. More particularly, the invention concerns an apparatus and method to increase call capacity through load balancing of signaling traffic workload in a wireless communication network radio control processor cluster with designated signaling links to subtending network elements communicating with one or more radio units.
2. Description of the Prior Art
Modern wireless communication networks, such as cellular telephone systems, personal communication systems, and the like, employ message signaling to provide operational, administrative and maintenance functionality in support of cell base stations that communicate with mobile (or non-mobile) network radio units. When a call is originated from, or terminated to, a radio unit located in a cell managed by a cell base station, a signaling system serving the base station routes signaling messages that enable the base station to perform the necessary call set-up, knock-down, hand-off and paging functions required for wireless cell-based communication. Signaling support may also be implemented for subscriber-specific intelligent network services, such as Call Forwarding, Call Waiting, Three-way Calling, Calling Line Identification and the like.
Existing wireless network signaling systems are implemented using programmed computers. These processing elements are referred to by various names, including “application processors” (APs). They are usually located in a mobile switching center (MSC), or in a base station controller (BSC) if such is in use in the mobile communication network. For convenience, the terms “application processor” and “AP” will be used hereinafter to refer to a signal processing component (described in more detail below) that processes signaling messages on behalf of cell base stations. The use of these terms is not intended to signify any particular architecture or commercial product offering.
In some wireless communication networks, such as the FLEXENT™ system from Lucent Technologies Inc., signaling functions are implemented on behalf of cell base stations by clusters of application processors arranged in a networked environment. The application processors act as network servers that maintain signaling links to the cell base stations, which function as subtending network client elements. Each application processor supports a fixed number of signaling links extending to several subtending cells (of different type and size).
An application processor cluster is a group of application processors that are mounted on a single frame or chassis having one or more power feeds. Often, there are two shelf racks on a frame, each with-a separate power feed, that hold two to four application processors, for a total of four to eight application processors per application cluster. Each application processor includes a CPU running multiple instances of the same radio control software (RCS), with each radio control software instance serving one or more cell base stations. By way of example, a single application processor may have anywhere from sixteen to forty-eight (or more) simultaneously executing radio control software instances, depending on the size of cells and the amount of signaling workload they produce.
In order to provide high levels of processor availability, as well as simplified administration and maintenance operations, prior art wireless communication networks typically assign each cell to a pair of application processors, both of which are co-located in the same application processor cluster, but which run off different power feeds. With respect to any given cell to which the application processor pair is connected, one of the application processors acts as the primary processor, and one of its radio control software instances runs in an active mode to handle all of the processing for the subtending cell. The other application processor acts as the secondary processor for the cell, and one of its radio control software instances is placed in a standby mode. This arrangement provides a degree of fail safety. In the event that the primary application processor fails or is taken off-line for maintenance, the secondary processor assumes the load, i.e., its radio control software instance becomes active. To facilitate a rapid transition between the primary and secondary application processors, two signaling links are permanently connected through a switching fabric between the cell and the pair of application processors providing service. Each signaling link typically comprises one DS
0
channel of a multi-channel (e.g., T
1
or E
1
) facility.
Although the foregoing redundancy is useful, it imposes a restriction on the distribution of application processors and the rated capacity of the radio control software running thereon. First, because the application processors are paired, a single processor fault (hardware/software/procedural) occurring during planned maintenance operations of the mate processor can interrupt service to all subtending cells. Second, with the paired processor approach, 40-50% of the CPU processing capacity must sit idle during normal operations. That is because each application processor is usually assigned to operate as the primary processor for one group of cells and as the secondary processor for another group of cells. Each application processor thus runs a plurality of primary/active radio control software instances (e.g., eight), and a plurality of secondary/standby radio control software instances (e.g., eight). The secondary/standby radio software instances represent the above-referenced 40-50% idle CPU capacity. This unused CPU capacity is reserved to run the workload of the mate processor in the processor pair in the event that the mate processor fails or is undergoing maintenance.
It will be appreciated that the prior art CPU distribution and recovery approach greatly restricts the rated capacity of the processor elements and the total system capacity for the network. With CPU availability limited by 40-50%, application processor overload can occur as a result of excessive message traffic and over-demand for signaling system functions. This can cause messaging delays and lower quality of service to wireless subscribers. Consider further that as wireless communication networks evolve and end user events migrate geographically, capacity needs and work loads will change. By way of example, some wireless communication network equipment providers plan to introduce modular base stations that will allow customers to change air interface technology and increase cell size and capacity over time. In such high mobility networks, the statically paired application processor approach of the prior art may not be able to sustain the rated processor capacity when more message traffic is generated during the life of a call.
As cells grow and capacity increases for one or more cells, the workload across the pair of processors may no longer be evenly distributed and may exceed the specified normal operating threshold (CPU utilization threshold) required to maintain rated call capacity on the terminating processors. Changes in call load (due to time of day, location or transient local end user/terminal events) can lead to one of the processors being over utilized and the other being under utilized.
Accommodating such workload imbalances in a conventional paired application processor configuration is impractical at best. To redistribute workload evenly, current systems require that the cells be taken off-line so that configuration changes can be made to the facility assignment on the switch connecting the application processors to the subtending cells. Of course, any configuration changes that interrupt service and increase maintenance procedures for wireless communicatio
Appiah Charles N.
Duft Walter W.
Lucent Technologies - Inc.
LandOfFree
Dynamic load balancing during message processing in a... 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 load balancing during message processing in a..., we encourage you to share that experience with our LandOfFree.com community. Your opinion is very important and Dynamic load balancing during message processing in a... will most certainly appreciate the feedback.
Profile ID: LFUS-PAI-O-3093494