Introduction to channel borrowing scheme cellular networks 1 | Page
Dec 13, 2014
Introduction to channel borrowing scheme cellular networks
By
Tanmoy Barman
HALDIA INSTITUTE OF TECHNOLOGY
1 | P a g e
Introduction
Advances in cellular mobile technology have engendered a new paradigm of computing, called mobile computing. The frequency spectrum allocated to this service is not sufficient with respect to enormous growth of mobile communication users. Tracking down a mobile user in a cellular network which is a collection of geometric areas called cells each serviced by a base station is the other concern to the designer. This service also needs another problem to be solved, that is the access of common information by mobile users. Other than the issues in existing technology, many other issues on developing technology need to be addressed. One main issue in cellular system design reduces to one of economics. Essentially we have a limited resource transmission spectrum that must be shared by several users. Unlike wired communications which benefits from isolation provided by cables, wireless users within close proximity of one another can cause significant interference to one another. To address this issue, the concept of cellular communications was introduced around in 1968 by researchers at AT&T Bell Labs.
2 | P a g e
Channel Allocation
There are several challenges in mobile cellular environment which is generally conceived as a collection of geometric areas called cells, each serviced by a base station (BS) located at its centre. A number of cells are again linked to a mobile switching centre (MSC) which also acts as a gateway of the cellular network to the existing wired network (shown in Figure 1). The problems in this area are clearly divided in two parts. Some of them are based on electronics and telecommunication and some of them are information based.
The basic concept being that a given geography is divided into polygons (hexagon) called cells. Each cell is allocated a portion of the total frequency spectrum. As users move into a given cell, they are then permitted to utilize the channel allocated to that cell. The virtue of the cellular system is that different cells can use the same channel given that the cells are separated by a minimum distance according to the system propagation characteristics; otherwise, intercellular or co-channel interference occurs. The minimum distance necessary to reduce co-channel interference is called the reuse distance. The reuse distance is defined as the ratio of the distance, D, between cells that can use the same channel without causing interference and the cell radius, R. Note that R is the distance from the center of a cell to the outermost point of the cell in cases when the cells are not circular.
3 | P a g e
Figure 1
A given radio spectrum is to be divided into a set of disjointed channels that can be used simultaneously while minimizing interference in adjacent channel by allocating channels appropriately (especially for traffic channels).
Channel allocation deals with the allocation of channels to cells in a cellular network. Once the channels are allocated, cells may then allow users within the cell to communicate via the available channels. Channels in a wireless communication system typically consist of time slots, frequency bands and/or CDMA pseudo noise sequences, but in an abstract sense, they can represent any generic transmission resource.
There are three major categories for assigning these channels to cells (or base-stations). They are
Fixed Channel Allocation, Dynamic Channel Allocation and Hybrid Channel Allocation.
4 | P a g e
Fixed Channel Allocation
Fixed Channel Allocation (FCA) systems allocate specific channels to specific cells. This allocation is static and cannot be changed. For efficient operation, FCA systems typically allocate channels in a manner that maximizes frequency reuse. Thus, in a FCA system, the distance between cells using the same channel is the minimum reuse distance for that system. The problem with FCA systems is quite simple and occurs whenever the offered traffic to a network of base stations is not uniform. Consider a case in which two adjacent cells are allocated N channels each. There clearly can be situations in which one cell has a need for N+k channels while the adjacent cell only requires N-m channels (for positive integers k and m). In such a case, k users in the first cell would be blocked from making calls while m channels in the second cell would go unused. Clearly in this situation of non-uniform spatial offered traffic, the available channels are not being used efficiently. FCA has been implemented on a widespread level to date(shown in Figure 2).
5 | P a g e
In FCA schemes, a set of channels is permanently allocated to each cell in the network.
If the total number of available channels in the system S is divided into sets, the minimum number of channel sets N required to serve the entire coverage area is related to the frequency reuse distance D as follows:
N = D2 / 3R2
Due to short term fluctuations in the traffic, FCA schemes are often not able to maintain high quality of service and capacity attainable with static traffic demands. One approach to address this problem is to borrow free channels from neighboring cells.
Figure 2
Dynamic Channel Allocation
In DCA schemes, all channels are kept in a central pool and are assigned dynamically to new calls as they arrive in the system. After each call is completed, the channel is returned to the central pool. It is fairly straightforward to select the most appropriate channel for any call based simply on current allocation and current traffic, with the aim of minimizing the interference. DCA scheme can overcome the problem of FCA
6 | P a g e
scheme. However, variations in DCA schemes center around the different cost functions used for selecting one of the candidate channels for assignment (shown in figure 3).
Figure 3
DCA schemes can be centralized or distributed.
The centralized DCA scheme involves a single controller selecting a channel for each cell;
The distributed DCA scheme involves a number of controllers scattered across the network (MSCs).
Centralized DCA schemes can theoretically provide the best performance. However, the enormous amount of computation and communication among BSs leads to excessive system latencies and renders centralized DCA schemes impractical. Nevertheless, centralized DCA schemes often provide a useful benchmark to compare practical decentralized DCA schemes.
7 | P a g e
Dynamic Channel Allocation (DCA) attempts to alleviate the problem mentioned for FCA systems when offered traffic is non-uniform. In DCA systems, no set relationship exists between channels and cells. Instead, channels are part of a pool of resources. Whenever a channel is needed by a cell, the channel is allocated under the constraint that frequency reuse requirements cannot be violated. There are two problems that typically occur with DCA based systems.
• First, DCA methods typically have a degree of randomness associated with them and this leads to the fact that frequency reuse is often not maximized unlike the case for FCA systems in which cells using the same channel are separated by the minimum reuse distance.
• Secondly, DCA methods often involve complex algorithms for deciding which available channel is most efficient. These algorithms can be very computationally intensive and may require large computing resources in order to be real-time.
Centralised DCS Scheme
For a new call, a free channel from the central pool is selected that would maximize the number of members in its co-channel set.
Minimize the mean square of distance between cells using the same channel.
Scheme Description
First Available (FA)
Among the DCA schemes the simplest one is the FA strategy. In F A, the first available channel within the reuse distance encountered during a channel search is assigned to the call.
The FA strategy minimizes the system computational time.
8 | P a g e
Locally Optimized Dynamic Assignment (LODA)
The channel selection is based on the future blocking probability in the vicinity of the cell where a call is initiated.
Scheme Description
Mean Square (MSQ),
The MSQ scheme selects the available channel that minimizes the mean square of the distance among the cells using the same channel.
1-clique This scheme uses a set of graphs, one for each channel, expressing the non co-channel interference structure over the whole service area for that channel.
Distributed DCA Scheme
Based on one of the three parameters:-
Co-channel distance
- co-channel cells in the neighborhood not using the channel.
- Sometimes adjacent channel interference taken in to account.
Signal strength measurement
- anticipated CIR above threshold.
9 | P a g e
Signal to noise interference ratio
- satisfy desired CIR ratio.
Hybrid Channel Allocation
HCA schemes are the combination of both FCA and DCA techniques. In HCA schemes, the total number of channels available for service is divided into fixed and dynamic sets. The fixed set contains a number of nominal channels that are assigned to cells as in the FCA schemes and, in all cases, are to be preferred for use in their respective cells. The dynamic set is shared by all users in the system to increase flexibility. Request for a channel from the dynamic set is initiated only when the cell has exhausted using all its channels from the fixed set (shown in figure 4).
Extra features:-
3:1 (fixed to dynamic), provides better service than fixed scheme for 50% traffic.
Beyond 50% fixed scheme perform better. For dynamic, with traffic load of 15% to 32%,
better results are found with HCA.
Example: When a call requires service from a cell and all of its nominal channels are busy, a channel from the dynamic set is assigned to the call.
10 | P a g e
Figure 4
Switching strategies
11 | P a g e
Comparison between FCA and DCA
FCA DCA
Performs better under heavy traffic
Low flexibility in channel assignment
Performs better under light/moderate traffic
Flexible channel allocation
Not always maximum
12 | P a g e
Maximum channel reusability
Sensitive to time and spatial changes
Not stable grade of service per cell in an interference cell group
High forced call termination probability
Suitable for large cell environment
Low flexibility
Radio equipment covers all channels assigned to the cell
Independent channel control
Low computational effort
Low call set up delay
Low implementation complexity
Complex, labor intensive frequency planning
Low signaling load
Centralized control
channel reusability
Insensitive to time and time spatial changes
Stable grade of service per cell in an interference cell group
Low to moderate forced call termination probability
Suitable in microcellular environment
High flexibility
Radio equipment covers the temporary channel assigned to the cell
Fully centralized to fully distributed control dependent on the scheme
High computational effort
Moderate to high call set up delay
Moderate to high implementation complexity
No frequency planning
Moderate to high signaling load
Centralized, distributed control depending on the scheme
13 | P a g e
Common Principles of Channel Allocation Schemes
The large array of possible channel allocation systems can become cumbersome. However, all channel allocation methods operate under simple, common principles. Throughout this report we have touched on three points which an efficient channel allocation scheme should address:
1.Channel allocation schemes must not violate minimum frequency reuse conditions.
2.Channel allocation schemes should adapt to changing traffic conditions.
3.Channel allocation schemes should approach (from above) the minimum frequency reuse constraints so as to efficiently utilize available transmission resources.
As the first requirement suggests, all channel allocation schemes adhere to condition 1. From a frequency reuse
14 | P a g e
standpoint, a fixed channel allocation system distributes frequency (or other transmission) resources to the cells in an optimum manner; i.e., common channels are separated by the minimum frequency reuse distance. Thus, a fixed channel allocation scheme perfectly satisfies condition 3 as well. However, a fixed allocation scheme does not satisfy condition 2.
Philosophically, any dynamic channel allocation scheme will meet the requirements of all of the
above three conditions to some degree. At the system architecture level dynamic channel allocation schemes may differ widely, but fundamentally, their only difference is in the degree to which they satisfy condition 3. Different DCA schemes attempt to satisfy condition 3 (in addition to conditions 1 and 2) by approaching the minimum frequency reuse constraint arbitrarily closely, and by doing so in as short a time period as possible. The above three conditions point to the fact that design of dynamic channel allocation schemes falls within the general class of optimization problems. Furthermore, since we can always assume that the available number of base stations is finite and the transmission resources will always be countable (due to FCC requirements if nothing else) then our problem can be reduced to the subclass of combinatorial optimization problems. As with all combinatorial optimization problems, there will exist a solution space and a cost function. A typical element of the solution space could be a particular layout of frequency channels among the base-stations. The cost function can be loosely characterized as the difference between the frequency reuse of an arbitrary solution and the frequency reuse of the optimized solution. The error associated with a non-optimized cost is realized as a future increased blocking probability or an otherwise unwarranted lack of channel availability. It is typically assumed that the solution to the wireless dynamic channel allocation problem is NP-complete [Yue, Cox - 1971]. The definition of np-completeness follows from the conjecture made in the late 1960's that there exists a class of combinatorial optimization problems of such inherent complexity that any algorithm, solving each instance of such a problem to optimality, requires a computational effort that grows super polynomially with the size of the problem. In the case of dynamic channel allocation, the complexity is generally attributed to the required inclusion of co-channel
15 | P a g e
interference in any analysis of dynamic channel allocation schemes. The author is aware of one published article to date offering an analytical method (approximate) for calculating the performance of dynamic channel allocation. Recently, several approximation techniques have been proposed as methods for solving condition 3 of the dynamic channel allocation problem. In particular there has been interest in applying simulated annealing techniques [Duque-Anton] and neural network methods to dynamic channel allocation.
Conclusion
This document has been briefly discussed about the static and dynamic allocation techniques on cellular networks. These techniques have been implemented in different areas and each technique has its advantages and disadvantages. Numbers of works are going on this field so far and furthermore many researches are going to make these techniques more stable. Growing up mobile technologies must need a reliable technique to cope up with this matter.
16 | P a g e
Bibliography
This document has been created with the help of
Wireless communication department paper Challenges of computing in mobile cellular environment—
a survey byS. DasBit*, S. Mitra
Wireless telecommunication Lab Document
17 | P a g e
18 | P a g e