Comprehensive Description of fixed and Dynamic Resource Allocation

Fixed resource allocation and dynamic resource allocation are two effective methods for transferring resources. Low-speed data terminals can use a high-speed line to achieve low-cost communication. This method is multiplexing. From the perspective of the allocation of transmission resources, there are two methods: fixed resource allocation and dynamic resource allocation.

1. Fixed Resource Allocation Method

The fixed allocation resource method is divided into "Time Division Multiplexing" TDM, Time Division Multiplexing) and "Frequency Division Multiplexing" OFDM, Frequency Division Multiplexing. Time division multiplexing is to distribute the transmission time of the line to each user in turn. Each user only sends and receives information using the line within the allocated time. In addition, when the user has no information to transmit during the allocated time, it cannot be used by other users, that is, it is only used for exclusive access during this time period. Figure 9-17 shows how four end users share resources in a time-sharing manner.

Here we assume that four end users use lines in turn. If no data is sent when a user is assigned, other users cannot use the user's time to send data. For example, in the first round, the data has arrived at the switching node. in Terminal 1, Terminal 2, and Terminal 4, the corresponding data is sent to A1, B1, and D1 respectively ), at this time, Terminal 3 does not send data, but the line is idle during this time period. In 2nd cases, Terminal 1 does not send data, and the line is idle. Terminal 2, Terminal 3, and Terminal 4 send data in their respective time periods, they are B2, C2, and D2. In the third round, Terminal 1, Terminal 2, and Terminal 4 do not send data, and the line is idle. It can be seen that in TDM multiplexing mode, only two interconnected users can establish point-to-point connections at any time. Sometimes there are many idle lines and the utilization of the lines is low.

Frequency Division Multiplexing refers to dividing the bandwidth resources of transmission lines, that is, bandwidth, into multiple sub-bands, and then allocating them to each user. Each user has a dedicated sub-channel for data transmission. The user uses this sub-channel for communication. When the user does not have information to transmit, it cannot be used by other users. Of course, because the total channel is divided into multiple sub-channels, the bandwidth of each sub-channel is much smaller, and the more terminals connected, the more sub-channels divided, the lower the sub-channel bandwidth, the lower the transmission rate.

In this example, we use four end users as an example. In this way, the total channel is divided into four small sub-channels, corresponding to Channels 1, 2, 3, and 4. Each sub-channel can freely arrange the data transmission time. For example, in a certain period of time, Terminal 1, Terminal 2, and Terminal 4 send data respectively. The corresponding data packets are A1, B1, and D1 respectively, when Terminal 3 does not send data, the corresponding sub-channel 3 is idle. In the next time period, all three sub-channels, except for Channel 1 corresponding to Terminal 1, send data, the corresponding data packets are B2, C2, and D2. From the principle, we can see that the OFDM multiplexing method improves the transmission efficiency compared with TDM, because each terminal can use its own sub-channel to send data at any time without waiting, however, there are still deficiencies in the line utilization rate, and the transmission rate will be greatly affected. After all, the bandwidth of each sub-channel is only a fraction of the total line bandwidth, or a few tenth. Based on the Communication Principles of the above two fixed resource allocation methods, we can see that the fixed resource allocation method has the disadvantages that the transmission bandwidth resources of the line cannot be fully utilized, the bandwidth resources allocated to a user cannot be used by other users even if they are idle.

2. Dynamic Resource Allocation Method

To overcome the disadvantages of the fixed resource allocation method, you can use the on-demand allocation method, that is, when the user needs to send data, it can be allocated to its line transmission resources, when no data is sent, no line transmission resources are allocated. This is the "Dynamic Resource Allocation" method, also known as "statistical time division multiplexing" STDM ). In this reuse method, the line resources can be used by other users, and the line transmission resources can be fully utilized to improve the line utilization. In this way, the dynamic allocation of resources is realized. Obviously, the utilization of the resource allocation method is much higher than that of the fixed allocation method.

In this resource allocation method, the data sent by each end user is first cached in the buffer, and sent in queue in sequence according to the sending time, instead of arranging the Sending team case according to the data sending sequence of a specific user. Each terminal can use the bandwidth of the entire line to send data at any time, instead of waiting for the data transmission period as long as the TDM multiplexing method, you do not have to transmit data in your own small channel, just like that of FMA. So how do all users differentiate the data sent by sharing lines? This is the role of the Group header. Because statistical time-division multiplexing does not identify the data transmitted by different users by location, rather it uses special tags, for example, the grouping header is used in group exchange to identify the data transmitted by different users, and the data sent from different users is differentiated based on the grouping header at the receiving end, then, the original data can be restored. In this multiplexing method, some readers may say that all users are transmitted in the same channel, and there will be a channel contention problem. In fact, it does not happen because data is not directly sent to the destination when it is sent from the end user to the STDM system, but cached in the buffer and then arranged to send the team instance in a unified manner.